• Selected
• Entire List

Abstract:
Using weekly data on advertising expenditures in various media and response data on awareness, consideration and choice, we test the hierarchy of effects hypothesis. Our empirical results, based on a simultaneous equations model with pooled parameters across brands, suggest that we can reject this hypothesis convincingly. Next, we consider a vector error correction model, again with pooled parameters, to see if there are dynamic effects of advertising. For the category under scrutiny, we find that most advertising effects exist for awareness, although at the same time there are effects for choice. Newspaper advertising turns out to be most influential.

advertising, awareness, consideration, choice

Abstract:
To comprehend the competitive structure of a market, it is important to understand the short-run and long-run effects of the marketing mix on market shares. A useful model to link market shares with marketing-mix variables, like price and promotion, is the market share attraction model. In this paper we put forward a representation of the attraction model, which allows for explicitly disentangling long-run from short-run effects. Our model also contains a second level, in which these dynamic effects are correlated with various brand and product category characteristics.Based on the findings in for example Nijs et al. (2001), we postulate the expected signs of these correlations. We fit our resultant Hierarchical Bayes attraction model to data on seven categories in two geographical areas. This data set spans a total of 50 brands. Our main finding is that, in absolute sense, the short-run price elasticity usually exceeds the long-run effect. Moreover, we find that the longrun price effects are strongly correlated with relative price and coupon intensity of a brand.

market shares, marketing mix, long-term effects, hierarchical bayes

Abstract:
To comprehend the competitive structure of a market, it is important to understand the short-run and long-run effects of the marketing mix on market shares. A useful model to link market shares with marketing-mix variables, like price and promotion, is the market share attraction model. In this paper we put forward a representation of the attraction model, which allows for explicitly disentangling long-run from short-run effects. Our model also contains a second level, in which these dynamic effects are correlated with various brand and product category characteristics.Based on the findings in for example Nijs et al. (2001), we postulate the expected signs of these correlations. We fit our resultant Hierarchical Bayes attraction model to data on seven categories in two geographical areas. This data set spans a total of 50 brands. Our main finding is that, in absolute sense, the short-run price elasticity usually exceeds the long-run effect. Moreover, we find that the longrun price effects are strongly correlated with relative price and coupon intensity of a brand.

market shares, marketing mix, long-term effects, hierarchical bayes

Abstract:
This paper contains a survey of the recent literature on the evaluation of direct marketing campaigns. We give an outline of the various stages included in such a campaign. Next, we review the statistical methods most frequently used and we review the general findings from using these methods.

direct marketing, evaluation, quantitative models, target selection

Abstract:
Coefficient Alpha, which is widely used in empirical research, estimates the reliability of a test consisting of parallel items. In practice it is difficult to compare values of alpha across studies as it depends on the number of items used. In this paper we provide a simple solution, which amounts to computing the confidence intervals of an alpha, as these intervals automatically account for differences across the numbers of items. We also give appropriate statistics to test for significant differences of alpha values across studies.

Cronbach's alpha, test reliability, confidence intervals

Abstract:
Ever more companies are recognizing the benefits of closed-loop supply chains that integrate product returns into business operations. IBM has been among the pioneers seeking to unlock the value dormant in these resources. We report on a project exploiting product returns as a source of spare parts. Key decisions include the choice of recovery opportunities to use, the channel design, and the coordination of alternative supply sources. We developed an analytic inventory control model and a simulation model to address these issues. Our results show that procurement cost savings largely outweigh reverse logistics costs and that information management is key to an efficient solution. Our recommendations provide a basis for significantly expanding the usage of the novel parts supply source, which allows for cutting procurement costs.

supply chain management, product recovery, reverse logistics, service management, inventory management

Abstract:
We develop two models to test hypotheses on the specific impact ofbrand and category characteristics on consumer stock-out responses.Our empirical results show that both characteristics are importantdeterminants. Consumers are more product loyal in hedonic productgroups than in utilitarian product groups and consumers are more brandloyal to high equity brands than to low equity brands. Brand loyaltyis especially strong for high equity brands in hedonic product groups.Our study also confirms findings from prior research on OOS reactions.Theoretical and managerial implications of the findings of the studyare discussed.

brand management, retailing, fast moving consumer goods, consumers, marketing-models

Abstract:
Market share attraction models are useful tools for analyzing competitive structures. The models can be used to infer cross-effects of marketing-mix variables, but also the own effects can be adequately estimated while conditioning on competitive reactions. Important features of attraction models are that they incorporate that market shares sum to unity and that the market shares of individual brands are in between 0 and 1. Next to analyzing competitive structures, attraction models are also often considered for forecasting market shares. The econometric analysis of the market share attraction model has not received much attention. Topics as specification, diagnostics, estimation and forecasting have not been thoroughly discussed in the academic marketing literature. In this chapter we go through a range of these topics, and, along the lines, we indicate that there are ample opportunities to improve upon present-day practice.

Market share attraction model, model selection, estimation, diagnostics, forecasting

Abstract:
We propose a new empirical representation of the Bass diffusion model, in order to estimate thethree key parameters, concerning innovation, imitation and maturity. The representation isbased on the notion that the observed data may temporarily deviate from the mean pathdetermined by the underlying hazard rate. Additionally, it rests on the idea that uncertaintyabout the cumulative process should be smaller, the closer it is to the start of the process and tothe level of maturity. Taking this into account, we arrive at an extension of the basicrepresentation proposed in Bass (1969), with an additional heteroskedastic error term. The typeof heteroskedasticity can be set by the modeler, as long as it obeys certain properties. Next, wediscuss the asymptotic theory for this new empirical model, that is, we focus on the properties ofthe estimators of the various parameters. We show that the parameters, upon standardizationby their standard errors, do not have the conventional asymptotic behavior. For practicalpurposes, it means that the t-statistics do not have an (approximate) t-distribution. Usingsimulation experiments, we address the issue how these findings carry over to practicalsituations. In a next set of simulation experiments, we compare the new representation withthat of Bass (1969) and Srinivasan and Mason (1986). We document that these last twoapproaches often seriously overestimate the precision of the parameter estimators. We alsoshed light on the effects of temporal aggregation and on the effects of a serious and persisentdeviation between the actual data and their mean. Finally, we consider the various empiricalrepresentations for a monthly series on installed ATMs.

Bass diffusion model, representation, estimation

Abstract:
We formulate a theoretical model in which we postulate that if a customers' behavior is perceived as not optimal, customers will adjust this behavior based on their current satisfaction and payment equity. Furthermore, customers will also include new experiences. In our empirical study we particularly investigate customer referrals and the amount of services purchased. Our results show positive effects of current satisfaction and payment equity on referrals, while also changes in satisfaction and payment equity affect customer referrals. With respect to the amount of services purchased, our estimation results reveal a positive significant effect of only changes in satisfaction.

dynamic modeling, satisfaction, customer relationships, preference updating

Abstract:
To understand the relevance of marketing efforts, it has become standard practice to estimatethe long-run and short-run effects of the marketing-mix, using, say, weekly scanner data. Acommon vehicle for this purpose is an econometric time series model. Issues that areaddressed in the literature are unit roots, cointegration, structural breaks and impulse responsefunctions. In this paper we summarize the most important concepts by reviewing all possibleempirical cases that can be encountered in practice using a prototypical model. We provideguidelines for practitioners, and illustrate these for a detailed workedout example.

dynamic effects, marketing mix, econometric time series models

Abstract:
A government’s ability to forecast key economic fundamentals accurately can affect business confidence, consumer sentiment, and foreign direct investment, among others. Government forecasts are subject to error, as can be seen by the frequent revisions that are made to initial, and even revised, official forecasts. A government forecast based on an econometric model is replicable, whereas one that is not based on an econometric model is non-replicable. Governments typically provide non-replicable forecasts of economic fundamentals, such as the inflation rate and real GDP growth rate. In this paper, we develop a model to generate one or more non-replicable government forecasts, examine the measurement errors contained in non-replicable government forecasts, compare replicable and non-replicable government forecasts using efficient estimation methods, and examine the accuracy of initial and updated (or revised) government forecasts. An empirical example to forecast economic fundamentals for Taiwan shows the relevance of the proposed methodological approach. The empirical analysis shows that replicable and non-replicable government forecasts can be distinctly different from each other, that efficient and inefficient estimation methods, as well as consistent and inconsistent covariance matrix estimates, can lead to significantly different outcomes, that government forecasts of economic fundamentals can differ markedly between initial and revised forecasts, and that alternative models and methods can lead to differences in the accuracy of government forecasts.

Abstract:
We examine the effect of relational constructs, such as satisfaction, trust and commitment on relationship performance (that is, positive word-of-mouth communication and the margin provided by each customer) of customers of an insurance company. A central issue concerns the effect of duration on the associations between relational constructs and relationship performance. Our empirical results provide strong evidence of duration dependent effects of satisfaction and trust, but we find only weak evidence of such effects on performance.

Relationship marketing, Performance, Relationship duration, Insurance industry

Abstract:
In this paper we put forward a brand choice model which incorporates responsiveness to marketing efforts as a form of structural heterogeneity. We introduce two latent segments of households. The households in the first segment are assumed to respond to marketing efforts while households in the second segment do not do so. Whether a specific household is a member of the first or the second segment at a specific purchase occasion is described by household-specific characteristics and characteristics concerning buying behavior. Households may switch between responsiveness states over time.We compare the in- and out-of-sample performance of our model with various versions of the MNL model. We conclude that, while using the smallest amount of parameters, our model outperforms all MNL variants on forecasting. This, together with the face validity of our parameter results, leads us to believe that incorporating responsiveness seems to be a worthwhile exercise.

Marketing-instrument effectiveness, structural heterogeneity, state dependence, multinomial logit, mixtures

Abstract:
Dividing forecasts of brand sales by a forecast of category sales, when they are generated from brand specific sales-response models, renders biased forecasts of the brands' market shares. In this paper we therefore propose an easy-to-apply simulation-based method which results in unbiased forecasts of the market shares. An illustration for five tuna fish brands emphasizes the practical relevance of the advocated method.

sales models, market shares, forecasting

Abstract:
The Bass (1969) diffusion theory often guides the construction of forecasting models for new product diffusion. To match the model with data, one needs to put forward a statistical model. This paper compares four empirical versions of the model, where two of these explicitly incorporate autoregressive dynamics. Next, it is shown that some of the regression models imply multi-step ahead forecasts that are biased. Therefore, one better relies on the simulation methods, which are put forward in this paper. An empirical analysis of twelve series (Van den Bulte and Lilien 1997) indicates that one-step ahead forecasts substantially improve by including autoregressive terms and that simulated two-step ahead forecasts are quite accurate.

diffusion, forecasting

Abstract:
The Bass (1969) diffusion theory often guides the construction of forecasting models for new product diffusion. To match the model with data, one needs to put forward a statistical model. This paper compares four empirical versions of the model, where two of these explicitly incorporate autoregressive dynamics. Next, it is shown that some of the regression models imply multi-step ahead forecasts that are biased. Therefore, one better relies on the simulation methods, which are put forward in this paper. An empirical analysis of twelve series (Van den Bulte and Lilien 1997) indicates that one-step ahead forecasts substantially improve by including autoregressive terms and that simulated two-step ahead forecasts are quite accurate.

diffusion, forecasting

Abstract:
Sales models are mainly used to analyze markets with afairly small number of items, obtained after aggregating to thebrand level. In practice one may require analyses at a moredisaggregate level. For example, brand managers may be interestedin a comparison across product attributes. For such an analysisthe number of relevant items in the product category make commonlyused sales models difficult to use as they would contain too manyparameters.In this paper we propose a new model, which allows for theanalysis of a market with many items while using only a moderatenumber of easily interpretable parameters. This is achieved bywriting the sales model as a Hierarchical Bayes model. In this waywe relate the marketing-mix effectiveness to item characteristicssuch as brand, package size, package type and shelf position. Inthis specification we do not have to impose restrictions on thecompetitive structure, as all items are allowed to have differentown and cross elasticities. The parameters in the model areestimated using Markov Chain Monte Carlo techniques.As a by-product this model allows to make predictions of sales levels and marketing-mix effectiveness of new to introduce itemsor of attribute changes. For example, one can assess the impact of changing the packaging from plastic to glass, on sales and price elasticity. Besides entering and changing products, our model also allows for items to leave the market.We consider the representation, specification and estimation ofthe model. We apply the model to a ketchup scanner data set with 23 items at the chain level. Our results indicate that the modelfits the sales of most items very well.

sales models, attribute data, SKU level analysis, hierarchical bayes, Markov Chain Monte Carlo

Abstract:
Changes in relationships are due to human actions. We assume that these human actions are functions of perceptions of a focal individual, but also the perceptions of other individuals who are part of the organizational and social environment. We hypothesize that perceptions based trust and perceptions of the structural environment individuals operate in affect relationship change more than the "actual" environment in which individuals operate. An empirically analysis shows the dynamic effects of perceptions on changes in two types of relationships, which are believed to be important in account management. We explore, 1, whether the levels of perceptions, and, 2, whether changes in perceptions affect relationship changes. For example, we consider the effects of the amount of trust as well as the change in the amount of trust one individual puts in another individual. We find that perceptions have more impact on relationship change than "actual" network variables have. Furthermore, the results show that it is useful to distinguish between level and change effects of perceptions.

relationship development, trust, dynamic cognitive social structures, network theory

Abstract:
The authors put forward a sales response model to explain the differences in immediate and dynamic effects of promotional prices and regular prices on sales. The model consists of a vector autoregression rewritten in error-correction format which allows to disentangle the immediate effects from the dynamic effects. In a second level of the model, the immediate price elasticities, the cumulative promotional price elasticity and the long-run regular price elasticity are correlated with various brand-speciffic and category-speciffic characteristics. The model is applied to seven years of data on weekly sales of 100 different brands in 25 product categories. We find many significant moderating effects on the elasticity of price promotions. Brands in categories that are characterized by high price differentiation and that constitute a lower share of budget are less sensitive to price discounts. Deep price discounts turn out to increase the immediate price sensitivity of customers. We also find significant effects for the cumulative elasticity. The immediate effect of a regular price change is often close to zero. The long-run effect of such a decrease usually amounts to an increase in sales. This is especially true in categories characterized by a large price dispersion, frequent price promotions and hedonic, non-perishable products.

sales, vector autoregression, marketing mix, promotional and regular price, short and long-term effects, hierarchical bayes

Abstract:
Indirect network effects are of prime interest to marketers because they affect the growth and takeoff of software availability for, and hardware sales of, a new product. While prior work on indirect network effects in the economics and marketing literature is valuable, these literatures show two main shortcomings. First, empirical analysis of indirect network effects is rare. Second, in contrast to the importance the prior literature credits to the chicken-and-egg paradox in these markets, the temporal pattern – which leads which? – of indirect network effects remains unstudied. Based on empirical evidence of nine markets, this study shows, among others, that: (1) indirect network effects, as commonly operationalized by prior literature, are weaker than expected from prior literature; (2) in most markets we examined, hardware sales leads software availability, while the reverse almost never happens, contradicting existing beliefs. These findings are supported by multiple methods, such as takeoff and time series analyses, and fit with the histories of the markets we studied. The findings have important implications for academia, public policy and management practice. To academia, it identifies a need for new, and more relevant, conceptualizations of indirect network effects. To public policy, it questions the need for intervention in network markets. To management practice, it downplays the importance of the availability of a large library of software for hardware technology to be successful.

Indirect Network Effects, New Product Growth, Takeoff, Chicken-and-Egg

Abstract:
Indirect network effects are of prime interest to marketers because they affect the growth and takeoff of software availability for, and hardware sales of, a new product. While prior work on indirect network effects in the economics and marketing literature is valuable, these literatures show two main shortcomings. First, empirical analysis of indirect network effects is rare. Second, in contrast to the importance the prior literature credits to the chicken-and-egg paradox in these markets, the temporal pattern - which leads which? - of indirect network effects remains unstudied. Based on empirical evidence of nine markets, this study shows, among others, that: (1) indirect network effects, as commonly operationalized by prior literature, are weaker than expected from prior literature; (2) in most markets we examined, hardware sales leads software availability, while the reverse almost never happens, contradicting existing beliefs. These findings are supported by multiple methods, such as takeoff and time series analyses, and fit with the histories of the markets we studied. The findings have important implications for academia, public policy and management practice. To academia, it identifies a need for new, and more relevant, conceptualizations of indirect network effects. To public policy, it questions the need for intervention in network markets. To management practice, it downplays the importance of the availability of a large library of software for hardware technology to be successful.

Indirect Network Effects, New Product Growth, Takeoff, Chicken-and-Egg

Abstract:
Allocating the proper amount of shelf space to stock keeping units [SKUs] is an increasingly relevant and difficult topic for managers. Shelf space is a scarce resource and it has to be distributed across a larger and larger number of items. It is in particular important because the amount of space allocated to a specific item has a substantial impact on the sales level of that item. This relation between shelf space and sales has been widely documented in the literature. However, besides the amount of space, the exact location of the SKU on the shelf is also an important moderator of sales. At the same time, the effectiveness of marketing instruments of an SKU may also depend on the shelf layout. In practice, retailers recognize that these dependencies exist. However, they often revert to rules of thumb to actually arrange their shelf layout. We propose a new model to optimize shelf arrangements in which we use a complete set of shelf descriptors. The goal of the paper is twofold. First of all, we aim to gain insight into the dependencies of SKU sales and SKU marketing effectiveness on the shelf layout. Second, we use these insights to improve the shelf layout in a practical setting. The basis of our model is a standard sales equation that explains sales from item-specific marketing-effect parameters and intercepts. In a Hierarchical Bayes fashion, we augment this model with a second equation that relates the effect parameters to shelf and SKU descriptors. We estimate the parameters of the two-level model using Bayesian methodology, in particular Gibbs sampling. Next, we optimize the total profit over the shelf arrangement. Using the posterior draws from our Gibbs sampling algorithm, we can generate the probability distribution of sales and profit in the optimization period for any feasible shelf arrangement. To find the optimal shelf arrangement, we use simulated annealing. This heuristic approach has proven to be able to effectively search an enormous solution space. Our results indicate that our model is able to fit and forecast the sales levels quite accurately. Next, when applying the simulated annealing algorithm to the shelf layout, we appear to be able to increase profits for all the stores analyzed. We compare our approach to commonly used shelf optimization rules of thumb. Most sensible rules of thumb also increase expected profits (although not as much as our optimization algorithm). In particular, it is beneficial to put high-margin items close to the beginning of the aisle (or the “racetrack"). Finally, we provide managerial implications and directions for further research.

Shelf Management, Sales Models, Hierarchical Bayes, Markov Chain Monte Carlo, Simulated Annealing

Abstract:
Marketing data appear in a variety of forms. An often-seen form is time-series data, like sales per month, prices over the last few years, market shares per week. Time-series data can be summarized in time-series models. In this chapter we review a few of these, focusing in particular on domains that have received considerable attention in the marketing literature. These are (1) the use of persistence modelling and (2) the use of state space models.

Time Series, Marketing, Persistence, State Space

Abstract:
Marketing literature has long recognized that brand price elasticity need not be monotonic and symmetric, but has yet to provide generalizable market-level insights on threshold-based price elasticity, asymmetric thresholds, and the sign and magnitude of elasticity transitions. This paper introduces smooth transition regression models to study threshold-based price elasticity of the top 4 brands across 20 fast-moving consumer good categories. Threshold-based price elasticity is found for 76% of all brands: 29% reflect historical benchmarkprices, 16% reflect competitive benchmarkprices, and 31% reflect both types of benchmarks. The authors demonstrate asymmetry for gains versus losses on three levels: the threshold size and the sign and the magnitude of the elasticity difference. Interestingly, they observe latitude of acceptance for gains compared to the historical benchmark, but saturation effects in most other cases. Moreover, category characteristics influence the extent and the nature of threshold-based price elasticity, while individual brand characteristics impact the size of the price thresholds. From a managerial perspective, the paper illustrates the sales, revenue, and margin implications for price changes typically observed in consumer markets.

kinked demand curve, smooth-transition regression models, time-series analysis, asymmetric price

Abstract:
Prior research on technology-intensive (TI) markets makes abstraction of the social context in which transactions take place. In contrast with this prior literature, the authors show that buyer-vendor transactions in TI markets are relationally and structurally embedded in an interfirm network. Their main premise is that buyers in TI markets prefer vendors with whom they can share a strong tie, and that in turn buyers want these vendors to share strong ties with their component manufacturers. This is an important addition to TI literature and to the on-going debate on the strength of ties in the sociology, management and marketing literatures. The authors also specifically consider how characteristics focal to TI markets, such as the know-how buyers possess or the pace of technological change they perceive, affect the extent to which buying behavior is relationally and structurally embedded. An empirical test in the computer network market shows good support for the developed theory.

technology-intensive markets, embeddedness, buying behavior, tie strength, conjoint analysis

Abstract:
A commonly applied modeling tool for the analysis of promotional effects onweekly sales data is a linear regression model. Usually, such a model includes0/1 dummy variables for promotions, where weeks with a promotion get a valueof 1. When these variables are included in a model with parameters which areconstant over time, the market researcher implicitly makes two important but ratherrestrictive assumptions. The first is that anytime a dummy variable takes a value of1 and the relevant parameter is significant, there is a non-zero effect of promotionon sales. The second is that this effect is constant across all weeks.In many practical cases however, one may conjecture that the effects of promo-tion are not constant over time. Therefore, we propose a new and rather parsimo-nious econometric model for the purpose of measuring the effects of promotions,while allowing for time-variation in these effects. The main idea is that promotionscan (but not necessarily) lead to positive and suddenly large values of sales in thesame week, and that they can perhaps lead to large negative values in the week there-after, if there is a, what is called, post-promotion dip. We discuss representation and interpretation of the model, and we outline the maximum likelihood parameterestimation method. Simulation results suggest that the estimation method is quitereliable and that the distribution of the estimator is approximately normal. Weillustrate the model in substantial detail on two sets of empirical data in order toindicate its practical usefulness

ales, promotions, time-varying effects, censored regression

Abstract:
An interlock between two firms occurs if the firms share one or more directors in their boards of directors. We explore the effect of interlocks on firm performance for 101 large Dutch firms using a large and new panel database. We use five different performance measures, and for each performance measure we design three different panel data models, where we allow the effect of the number of interlocks to be linear, quadratic or square root, either with or without lags. Based on all results we conclude that current interlocks can have a negative effect on future firm performance. We show that this negative effect is jointly established by (1) interlocking directors being too busy and (2) by directors being members of a homogenous upper class group.

interlocks, firm performance

Abstract:
This paper develops a return forecasting methodology that allows for instability in the relationship between stock returns and predictor variables, for model uncertainty, and for parameter estimation uncertainty. The predictive regression specification that is put forward allows for occasional structural breaks of random magnitude in the regression parameters, and for uncertainty about the inclusion of forecasting variables, and about the parameter values by employing Bayesian Model Averaging. The implications of these three sources of uncertainty, and their relative importance, are investigated from an active investment management perspective. It is found that the economic value of incorporating all three sources of uncertainty is considerable. A typical in vestor would be willing to pay up to several hundreds of basis points annually to switch from a passive buy-and-hold strategy to an active strategy based on a return forecasting model that allows for model and parameter uncertainty as well as structural breaks in the regression parameters.

Stock return predictability, model uncertainty, Bayesian model averaging, structural breaks, portfolio selection

Abstract:
In this paper we develop a new semi-parametric model for conditional correlations, which combines parametric univariate GARCH-type specifications for the individual conditional volatilities with nonparametric kernel regression for the conditional correlations. This approach not only avoids the proliferation of parameters as the number of assets becomes large, which typically happens in conventional multivariate conditional volatility models, but also the rigid structure imposed by more parsimonious models, such as the dynamic conditional correlation model. An empirical application to the 30 Dow Jones stocks demonstrates that the model is able to capture interesting asymmetries in correlations and that it is competitive with standard parametric models in terms of constructing minimum variance portfolios and minimum tracking error portfolios.

Multivariate GARCH, dynamic conditional correlation, kernel regression, minimum variance portfolio, tracking error minimization

Abstract:
Market share models for weekly store-level data are useful to understand competitive structures by delivering own and cross price elasticities. These models can however not be used to examine which brands lose share to which brands during a specific period of time. It is for this purpose that we propose a new model, which does allow for such an examination. We illustrate the model for two product categories in two markets, and we show that our model has validity in terms of both in-sample fit and out-of-sample forecasting. We also demonstrate how our model can be used to decompose own and cross price elasticities to get additional insights into the competitive structure.

Competitive structure, elasticity decomposition, market shares, share-switching, store-level scanner data

Abstract:
We propose a consistent utility-based framework to jointly explain a household's decisions on purchase incidence, brand choice and purchase quantity. The approach differs from other approaches, currently available in the literature, as it is able to take into account consumption dynamics. In the model, households derive utility from consumption, and they relate their purchase behavior to consumption planning. We illustrate our model for yogurt purchases, and show that our model yields important additional insights. One such insight is that the reservation price of households is not fixed, but depends on the available inventory stock. Furthermore, we find that promotional activities increase sales through more purchases in the product category and brand switching, but the effect through larger purchase quantities is limited.

purchase incidence, brand choice, purchase quantity, consumption, utility maximization

Abstract:
In this paper we empirically investigate various benefits and costs associated with broker characteristics of individuals who operate in the account management system of financial service providers. We narrow our focus to broker positions in two specific task-specific knowledge networks that facilitate account management. We study the effect of broker positions on the contribution of individuals to organizational performance. We measure such a contribution by measuring the perceptions of others concerning a particular individual. We also explore how certain personal costs are associated with these task-specific broker positions. More specifically, we explore how these positions affect role ambiguity and role conflict, as self-perceived by that particular individual. To test the hypothesized effects we collect data for a network consisting of 55 individuals. We conclude with stating that service specification broker positions benefit organizations, but service delivery broker positions are detrimental to an organization and that they also invoke personal costs.

social networks, task-specific broker positions, role stress, account management

Abstract:
Using only aggregate sales data, the model we propose decomposes the diffusion processes of the respective technological generations and tests if different technological generations have different diffusion parameters. It also estimates the location of the generational transition from the old to the new technology. We develop a routine to test whether the maturation point of the old generation occurs before or after the transition to a new technological generation. Finally, we show that, when the aggregate sales data are generated by multiple technological generations, our model does better in forecasting than a single-regime Bass model.

technological generations, regime-switching models, diffusion modeling

Abstract:
Using only aggregate sales data, the model we propose decomposes the diffusion processes of the respective technological generations and tests if different technological generations have different diffusion parameters. It also estimates the location of the generational transition from the old to the new technology. We develop a routine to test whether the maturation point of the old generation occurs before or after the transition to a new technological generation. Finally, we show that, when the aggregate sales data are generated by multiple technological generations, our model does better in forecasting than a single-regime Bass model.

technological generations, regime-switching models, diffusion modeling

Abstract:
Marketing problems sometimes concern the analysis of dichotomous variables, like for example ``buy'' and ``not buy'' and ``respond'' and ``not respond''. It can happen that one outcome strongly outnumbers the other, for example when many households do not respond (to a direct mailing, for example). Standard econometric methods would imply the collection of many data to obtain precise estimates and this can be rather costly. To cut back costs, we propose to implement a non-random sampling scheme and to correct for the subsequent sample selection bias in the econometric model. In this paper we put forward the relevant method, which does not lead to a loss in precision. Our illustration suggests an opportunity to collect 60\% less data points.

Outcome-dependent sampling, sample size, survey design, binary outcomes, logit model

Abstract:
We propose a two-stage MRQAP to analyze dynamic network data, within the framework of an equilibrium-correction (EC) model. Extensive simulation results indicate practical relevance of our method and its improvement over standard OLS. An empirical illustration additionally shows that the EC model yields interpretable parameters, in contrast to an unrestricted dynamic model.

structural autocorrelation, two-stage equilibrium model, consistent accuracy, cognitive social structures, network centrality

Abstract:
This paper puts forward a method to estimate average economic growth, and its associated confidence bounds, which does not require a formal decision on potential unit root properties. The method is based on the analysis of either difference-stationary or trend-stationary time series models, implementing the robust bootstrapping procedure advocated in Romano and Wolf (2001). Simulation evidence indicates the practical relevance of the method. It is illustrated on quarterly post-war US industrial production.

Growth, Unit root, Robust testing

Abstract:
To examine cross-country diffusion of new products, marketing researchers have to rely on a multivariate product growth model. We put forward such a model, and show that it is a natural extension of the original Bass (1969) model. We contrast our model with currently in use multivariate models and we show that inference is much easier and interpretation is straightforward. In fact, parameter estimation can be done using standard commercially available software. We illustrate the benefits of our model relative to other models in simulation experiments. An application to a three-country CD sales series shows the merits of our model in practice.

Diffusion, international marketing, econometric models

Abstract:
We propose a new method to model consumers' consideration and choice processes. We develop a parsimonious probit type model for consideration and a multinomial probit model for choice, given consideration. Unlike earlier models of consideration ours is not prone to the curse of dimensionality, while we allow for very general structures of unobserved dependence in consideration among brands. In addition, our model allows for state dependence and marketing mix effects on consideration.Unique to this study is that we attempt to establish the validity of existing practice to infer consideration sets from observed choices in panel data. To this end, we use data collected in an on-line choice experiment involving interactive supermarket shelves and post-choice questionnaires to measure the choice protocol and stated consideration levels. We show with these experimental data that underlying consideration sets can be successfully retrieved from choice data alone and that there is substantial convergent validity of the stated and inferred consideration sets. We further find that consideration is a function of point-of-purchase marketing actions such as display and shelf space, and of consumer memory for recent choices.Next, we estimate the model on IRI panel data. We have three main results. First, compared with the single-stage probit model, promotion effects are larger and are inferred with smaller variances when they are included in the consideration stage of the two-stage model. Promotion effects are significant only in the two-stage model that includes consideration, whereas they are not in a single-stage choice model. Second, the price response curves of the two models are markedly diferent. The two-stage model offers a nice intuition for why promotional price response is different from regular price response. In addition and consistent with intuition, the two-stage model also implies that merchandizing has more effect on choice among those who did not buy the brand before than among those who already did. It is explained why a single-stage model does not harbor this feature. In fact, the single-stage model implies the opposite for smaller or more expensive brands. Third, we find that the consideration of brands does not covary greatly across brands once we take account of observed effects. Managerial implications and future research are also discussed.

Consideration, choice, probit models

Abstract:
The abundance of highly disaggregate data (e.g., at 5 second intervals) raises the question of the optimal data interval to estimate advertising carryover. The literature assumes that 1) the optimal data interval is the inter-purchase time, 2) too disaggregate data causes a disaggregation bias and 3) recovery of true parameters requires assumption of the underlying advertising process. In contrast, we show that 1) the optimal data interval is the inter-exposure time. 2) Too disaggregate data does not cause any disaggregation bias, and 3) recovery of true parameters does not require assumption of the advertising process but only data at the inter-exposure time. These results hold for any linear dynamic model linking sales with current and past advertising.

Advertising carryover, Data Interval, Econometric Models

Abstract:
We propose that weather conditions can influence consumers’ engagement in lottery play. A longitudinal study on the extent of lottery play in Belgium shows that lottery expenditures are indeed higher after reduced exposure to sunshine, even after controlling for people’s inertia, time-varying characteristics of the game, and deterministic seasonal components. The results of a first laboratory study are consistent with these findings, and establish a link between lottery play and negative mood. Subsequent experiments provide evidence that depletion due to active mood regulation attempts, rather than mood repair, is the underlying process for the link between bad weather and lottery play.

lottery, lottery play, mood repair, gambling, weather-induced

Abstract:
It is conceivable that the "whether to buy" and "how much to buy" decisions in the purchasing process of households are influenced by the inventory process. In this paper we therefore put forward a model for consumption, where we rely on established economic theory. We incorporate this model in a model for purchase behavior. Our consumption specification, which is derived from utility maximization principles, is more flexible than an ad hoc approach, which has recently been proposed in the literature. We illustrate our model for yogurt purchases, and show that our model yields important additional and useful insights. One such insight is that promotion anticipation behavior turns out not only to occur in the purchasing process, but also in the consumption process.

consumption function, inventory, utility maximization, promotion anticipation

Abstract:
We present a model that integrates the contradicting Burtian and Krackhardtian broker theories to explain effectiveness of brokering for individuals within account management organizations. Using data on a network of 55 individuals in a financial account management organization, we test how brokerage of different resource relationships and Simmelian trust relationships affect individual effectiveness. We find that although brokering in ‘specification’ processes enhances effectiveness, it harms to broker in ‘delivery’ processes. Furthermore, brokers of Simmelian trust relationships appear to face more diverse role expectations, which causes role ambiguity that reduces effectiveness. These results have implications for account management organization.

social networks, broker theory, account management, simmelian ties, network constraint

Abstract:
We propose a dynamic direct mailing response model with competitive effects, where purchase and promotion history are incorporated. We then map the dynamic competitive interactions amongst the firms sending the mailings. We investigate the short- and long-run impact of a direct mailing on the revenues of the firm sending the mailing and on the revenues of its competitors. The model accounts for unobserved heterogeneity across households. We estimate the model in the charitable giving setting, as sending direct mailings represents a large part of charitable fundraising activity. Households often receive direct mailings of different charities within a short period of time and competition is highly relevant. We construct a unique database by merging the databases of three large charity organizations in the Netherlands. This results in household level data on the direct mailings received and the donations made by each household to each charity. Our results show that charitable direct mailings are short-run complements, that is, the direct mailings tend to increase the total pie that is divided among the charities. At the same time, the charitable direct mailings are long-run substitutes. In the long run they fight for a piece of the pie that households have available for charitable giving.

Dynamics, Competition, Direct Mailings

Abstract:
Estimation results in linear regression models are sometimes in contrast with what was expected on the basis of a certain set of hypotheses or theory, in the sense that one or more parameters have the "wrong sign". One could be inclined to think that this is due to collinearity across explanatory variables, suggesting one should leave out one or more of the collinear variables. In this note we show that this is not a valid approach. Additionally, we show that "wrong signs" can occur because of correlations between included and omitted variables, so that "wrong signs" may occur if the model is not correctly specified. That is, if we find 'wrong signs" we should start questioning our model choice, not the data.

misspecification, collinearity, parameter estimation

Abstract:
This paper discusses the practical usefulness of seasonally adjusted time series data. Aspects of seasonal adjustment are considered, and the relevance of adjusted data for economic modelling is examined. One recommendation which emerges from the discussion is that the adjusted data should be presented together with their estimated standard errors. Another is that it is perhaps better not to seasonally adjust at all.

Seasonal adjustment, time series

Abstract:
The accuracy of real-time forecasts of macroeconomic variables that are subject to revisions may crucially depend on the choice of data used to compare the forecasts against. We put forward a flexible time - varying parameter regression framework to obtain early estimates of the final value of macroeconomic variables based upon the initial data release that may be used as actuals in current forecast evaluation. We allow for structural changes in the regression parameters to accommodate benchmark revisions and definitional changes, which fundamentally change the statistical properties of the variable of interest, including the relationship between the final value and the initial release. The usefulness of our approach is demonstrated through an empirical application comparing the accuracy of forecasts of US GDP growth rates from the Survey of Professional Forecasters and the Greenbook.

Data revision, forecast evaluation, parameter uncertainty, Bayesian estimation, structural breaks

Abstract:
We examine how new products are priced over time, where we particularly look at sharp decreases in prices. New durable products like fashion, apparel, and videogames often show a significant price cut some time after the product’s introduction. We call this a price landing and we examine its drivers. Theory predicts that competitive effects or underperforming sales are drivers for such price landings. To our knowledge, however, a systematic empirical study of price landing is unavailable. To examine the drivers of significant price cuts of a new product, we consider a rich dataset concerning sales and prices of 1195 newly released videogames. Prior literature suggests that own sales, competitive sales, competitive prices or simply time could be such drivers. In this paper we put these suggestions to an empirical test. We put forward a mixture model that covers a set of pricing equations with the price landing moment and its speed as key parameters. Second, in a hierarchical model we explain the apparent heterogeneity across the products. Our main finding is that it is not sales thresholds but competition and time itself that makes managers decide to seriously cut prices.

pricing, pricing models, new products

Abstract:
We examine how new products are priced over time, where we particularly look at sharp decreases in prices. New durable products like fashion, apparel, and videogames often show a significant price cut some time after the product’s introduction. We call this a price landing and we examine its drivers. Theory predicts that competitive effects or underperforming sales are drivers for such price landings. To our knowledge, however, a systematic empirical study of price landing is unavailable. To examine the drivers of significant price cuts of a new product, we consider a rich dataset concerning sales and prices of 1195 newly released videogames. Prior literature suggests that own sales, competitive sales, competitive prices or simply time could be such drivers. In this paper we put these suggestions to an empirical test. We put forward a mixture model that covers a set of pricing equations with the price landing moment and its speed as key parameters. Second, in a hierarchical model we explain the apparent heterogeneity across the products. Our main finding is that it is not sales thresholds but competition and time itself that makes managers decide to seriously cut prices.

pricing, pricing models, new products

Abstract:
We examine how new products are priced over time, where we particularly look at sharp decreases in prices. New durable products like fashion, apparel, and videogames often show a significant price cut some time after the product's introduction. We call this a price landing and we examine its drivers. Theory predicts that competitive effects or underperforming sales are drivers for such price landings. To our knowledge, however, a systematic empirical study of price landing is unavailable. To examine the drivers of significant price cuts of a new product, we consider a rich dataset concerning sales and prices of 1195 newly released videogames. Prior literature suggests that own sales, competitive sales, competitive prices or simply time could be such drivers. In this paper we put these suggestions to an empirical test. We put forward a mixture model that covers a set of pricing equations with the price landing moment and its speed as key parameters. Second, in a hierarchical model we explain the apparent heterogeneity across the products. Our main finding is that it is not sales thresholds but competition and time itself that makes managers decide to seriously cut prices.

pricing, pricing models, new products

Abstract:
Direct mailing is the main tool that charities employ for fundraising. With increasing amounts of soliciting mailings and with the best donators receiving more mailings as a result of target selection, irritation might increase. As a result, such irritation could cause individuals to donate less, and hence reduce revenues for charities. We develop a conceptual model, which relates donating behavior to irritation and to mailing frequencies. We consider mailing frequencies relative to a reference point, which we call the maximum acceptance level. Furthermore, we allow for asymmetric effects of positive and negative differences with this maximum acceptance level, and hence we consider the effects of receiving excessive and acceptable amounts of mailings. To test our model empirically, we conduct a survey on charitable direct mailings and donating behavior among 213 respondents. We find that too many mailings do indeed lead to irritation, and that such irritation reduces annual donations.

DM, Irritation, Junk Mail, Direct Mail

Abstract:
Market share models for weekly store-level data are useful to understand competitive structuresby delivering own and cross price elasticities. These models can however not be used toexamine which brands lose share to which brands during a specific period of time. It is for thispurpose that we propose a new model, which does allow for such an examination. We illustratethe model for two product categories in two markets, and we show that our model has validity interms of both in-sample fit and out-of-sample forecasting. We also demonstrate how our modelcan be used to decompose own and cross price elasticities to get additional insights into thecompetitive structure.

competitive structure, elasticity decomposition, market shares, share-switching, store-level scanner data

Abstract:
We study the performance of sales forecasts which linearly combine model-based forecasts and expert forecasts. Using a unique and very large database containing monthly model-based forecasts for many pharmaceutical products and forecasts given by thirty-seven different experts, we document that a combination almost always is most accurate. When correlating the specific weights in these "best" linear combinations with experts' experience and behaviour, we find that more experience is beneficial for forecasts for nearby horizons. And, when the rate of bracketing increases the relative weights converge to a 50%-50% distribution, when there is some slight variation across forecasts horizons.

model-based forecasts, experts forecast, combining forecasts

Abstract:
Marketing literature has long recognized that price response need not be monotonic and symmetric, but has yet to provide generalizable market-level insights on reference price type, asymmetric thresholds and sign and magnitude of elasticity transitions. In this paper, we introduce smooth transition models to study reference-based price response across 25 fast moving consumer good categories. Our application to 100 brands shows that 77% demonstrate reference-based price response, of which 36% reflects historical reference prices, 31% reflects competitive reference prices, and 33% reflects both types of reference prices. This reference-based price response shows asymmetry for gains versus losses on three levels: the threshold size, the sign and the magnitude of the elasticity difference. For historical reference prices, the threshold size is larger for gains (20%) than for losses (12%) and the assimilation/contrast effects for gains (-0.41) are smaller than the saturation effects for losses (0.81). For competitive reference prices, the threshold size is smaller for gains (3%) than for losses (16%), and the saturation effects are larger for gains (0.33) than for losses (0.15). These results are moderated by both brand and category characteristics that affect reference price accessibility and diagnosticity. Historical reference prices more often play a role for national brands, for planned purchases and in inexpensive categories with low price volatility and high purchase frequency. When price discounting, high-share brands face larger latitudes of acceptance. When raising prices, saturation effects set in later for brands with high price volatility and for categories with high price spread and for planned purchases. As for competitive reference prices, saturation effects set in later for expensive brands with high price volatility and in categories with lower price volatility, higher price spread and higher concentration. Sales, revenue and margin implications are illustrated for price changes typically observed in consumer markets.

kinked demand curve, smooth-transition regression models, competitive versus historical reference prices, asymmetric price thresholds, saturation versus assimilation/contrast effects, empirical generalizations

Abstract:
Duration intervals measure the dynamic impact of advertising on sales. More precise, the p per cent duration interval measures the time lag between the advertising impulse and the moment that p per cent of its effect has decayed. In this paper, we derive an expression for the duration interval for a general dynamic model linking sales to advertising. Additionally, and this is themain novelty of the paper, we put forward a method to provide confidence bounds around the estimated duration interval. An illustration to real-life data emphasizes its usefulness.

advertising effects, duration interval, simulation, marketing

Abstract:
The authors investigate the impact of direct-response commercials on incoming calls at a national call center. To this end, the authors analyze the data of a fast service for repairs of (parts of) a durable consumption good in Flanders, Belgium. The authors have access to data at the 15 minute interval covering 30 months in which 5172 radio commercials were broadcasted on six radio stations at various times of the day and at with differing commercial lengths. Their model is a two-level model, where the first-level estimates of the short-run and long-run effects are correlated with various aspects of the commercial is the second level. Their main conclusion is that GRPs are the key drivers of the effectiveness of commercials.

advertising effectiveness, two-level model, advertising response, long-run elasticity, short-run effects, HF5837

Abstract:
We ask various experts, who produce sales forecasts that can differ from earlier received model-based forecasts, what they do and why they do so. A questionnaire with a range of questions was completed by no less than forty-two such experts who are located in twenty different countries. We correlate the answers to these questions with actual behavior of the experts. Our main findings are that experts have a tendency to double count and to react strongly to recent volatility in sales data. Also, experts who feel more confident give forecasts that differ most from model-based forecasts.

model forecasts, expert forecasts, decision making, stated behavior

Abstract:
Seasonal adjustment methods transform observed time series data into estimated data, where these estimated data are constructed such that they show no or almost no seasonal variation. An advantage of model-based methods is that these can provide confidence intervals around the seasonally adjusted data. One particularly useful time series model for seasonal adjustment is the basic structural time series (BSM) model. The usual premise of the BSM is that the variance of each of the components is constant. In this paper we address the possibility that the variance of the trend component in a macroeconomic time series in some way depends on the business cycle. One reason for doing so is that one can expect that there is more uncertainty in recession periods. We extend the BSM by allowing for a business-cycle dependent variance in the level equation. Next we show how this affects the confidence intervals of seasonally adjusted data. We apply our extended BSM to monthly US unemployment and we show that the estimated confidence intervals for seasonally adjusted unemployment change with past changes in the oil price.

Abstract:
In contrast to, for example, books and compact discs, the number of complex services offered on the Internet is still small. A good example of such a service concerns mortgage loans. The decision-making process differs for complex services in that they have an extra intermediate step of `indication of interest'. The web site is (1) visited and searched for information, subsequently (2) a request for the service is made, which may lead to (3) a purchase. This difference in the buying process and the complexity of the decision-making process, requires afurther investigation on purchasing complex services on the Internet. We therefore focus on online purchases of complex services, paying special attention to the determinants of a purchase of such services. To this end, we acquired a unique data set from an online Dutch financial service provider, which offers services like mortgage loans and insurances on the Internet. This data contains, besides clickstream data, also data on user-specific information like demographics. We also obtained information on whether the request for the service re-sulted in a purchase. Search behavior, product familiarity and trust appear to be useful determinants of purchase of complex services. Direct managerial applications of our model include the ability to identify customer characteristics of successful applicants, and subsequently the selection of customers.

online purchasing behavior, search behavior, trust, decision support system, internet

Abstract:
Charities mainly rely on direct mailings to attract the attention of potential donators. Individuals may feel irritated by these mailings, in particular when they receive many mailings. We study the consequences of perceived irritation on stated behavior and on actual behavior. Target selection by charities likely results in good donators receiving many mailings and hence they might also be most irritated. Therefore, irritation with direct mailings might be endogenously determined. To create exogenous variation in irritation, we design a unique controlled field experiment in cooperation with five of the largest charities in the Netherlands. Our analysis reveals that direct mailings do result in irritation, but surprisingly this affects neither stated nor actual donating behavior.

direct marketing, irritation, charity donations, field experiment

Abstract:
No abstract available.

Abstract:
Estimation results in linear regression models are sometimes in contrast with what was expected on the basis of a certain set of hypotheses or theory, in the sense that one or more parameters have the "wrong sign". One could be inclined to think that this is due to collinearity across explanatory variables, suggesting one should leave out one or more of the collinear variables. In this note we show that this is not a valid approach. Additionally, we show that "wrong signs" can occur because of correlations between included and omitted variables, so that "wrong signs" may occur if the model is not correctly specified. That is, if we find 'wrong signs" we should start questioning our model choice, not the data.

misspecification, collinearity, parameter estimation

Abstract:
Charities mainly rely on direct mailings to attract the attention of potential donators. Individuals may feel irritated by these mailings, in particular when they receive many mailings. We study the consequences of perceived irritation on stated behavior and on actual behavior. Target selection by charities likely results in good donators receiving many mailings and hence they might also be most irritated. Therefore, irritation with direct mailings might be endogenously determined. To create exogenous variation in irritation, we design a unique controlled field experiment in cooperation with five of the largest charities in the Netherlands. Our analysis reveals that direct mailings do result in irritation, but surprisingly this affects neither stated nor actual donating behavior.

direct marketing, irritation, charity donations, field experiment

Abstract:
We discuss a method to estimate the confidence bounds for average economic growth, which is robust to misspecification of the unit root property of a given time series. We derive asymptotic theory for the consequences of such misspecification. Our empirical method amounts to an implementation of the subsampling procedure advocated in Romano and Wolf (Econometrica, 2001, Vol. 69, p. 1283). Simulation evidence supports the theory and it also indicates the practical relevance of the subsampling method. We use quarterly postwar US industrial production for illustration and we show that non-robust approaches rather lead to different conclusions on average economic growth than our robust approach.

Abstract:
Marketing literature has long recognized that price response need not be monotonic and symmetric, but has yet to provide generalizable market-level insights on reference price type, asymmetric thresholds and sign and magnitude of elasticity transitions. In this paper, we introduce smooth transition models to study reference-based price response across 25 fast moving consumer good categories. Our application to 100 brands shows that 77% demonstrate reference-based price response, of which 36% reflects historical reference prices, 31% reflects competitive reference prices, and 33% reflects both types of reference prices. This reference-based price response shows asymmetry for gains versus losses on three levels: the threshold size, the sign and the magnitude of the elasticity difference. For historical reference prices, the threshold size is larger for gains (20%) than for losses (12%) and the assimilation/contrast effects for gains (-0.41) are smaller than the saturation effects for losses (0.81). For competitive reference prices, the threshold size is smaller for gains (3%) than for losses (16%), and the saturation effects are larger for gains (0.33) than for losses (0.15). These results are moderated by both brand and category characteristics that affect reference price accessibility and diagnosticity. Historical reference prices more often play a role for national brands, for planned purchases and in inexpensive categories with low price volatility and high purchase frequency. When price discounting, high-share brands face larger latitudes of acceptance. When raising prices, saturation effects set in later for brands with high price volatility and for categories with high price spread and for planned purchases. As for competitive reference prices, saturation effects set in later for expensive brands with high price volatility and in categories with lower price volatility, higher price spread and higher concentration. Sales, revenue and margin implications are illustrated for price changes typically observed in consumer markets.

kinked demand curve, smooth-transition regression models, competitive versus historical reference prices, asymmetric price thresholds, saturation versus assimilation/contrast effects, empirical generalizations

Abstract:
Marketing problems sometimes concern the analysis of dichotomous variables, like for example ``buy'' and ``not buy'' and ``respond'' and ``not respond''. It can happen that one outcome strongly outnumbers the other, for example when many households do not respond (to a direct mailing, for example). Standard econometric methods would imply the collection of many data to obtain precise estimates and this can be rather costly. To cut back costs, we propose to implement a non-random sampling scheme and to correct for the subsequent sample selection bias in the econometric model. In this paper we put forward the relevant method, which does not lead to a loss in precision. Our illustration suggests an opportunity to collect 60\% less data points.

Outcome-dependent sampling, sample size, survey design, binary outcomes, logit model

Abstract:
This paper reviews various recent approaches to cointegration analysis of seasonal time series. In addition to the usual decisions concerning data transformations and univariate time series properties, it is necessary to decide how seasonal variation is included in the multivariate model and how standard cointegration methods should accordingly be modified. Seasonal cointegration and periodic cointegration methods are discussed, as are some of their recent refinements. An overview of further research topics is also provided.

Abstract:
GARCH models and Stochastic Volatility (SV) models can both be used to describe unobserved volatility in asset returns. We consider the issue of testing a GARCH model against an SV model. For that purpose, we propose a new and parsimonious GARCH-t model with an additional restricted moving average term, which can capture SV model properties. We discuss model representation, parameter estimation, and our simple test for model selection. Furthermore, we derive the theoretical moments and the autocorrelation function of our new model. We illustrate our model and test for nine daily stock-return series.

C22, C52, GARCH, model selection, stochastic volatility

Abstract:
Philip Franses' students often ask how many data points they should collect when they have a project involving seasonality. His response is that there is no definite number. Factors to be considered are what is meant by enough data, whether seasonality is constant or changing, and which seasonal models are being used. If seasonality is slowly evolving, time-dependent as well as periodic models may be useful.

Abstract:
The abundance of highly disaggregate data (e.g., at 5-second intervals) raises the question of the optimal data interval to estimate advertising carryover. The literature assumes that 1) the optimal interval is the inter-purchase time, 2) too disaggregate data causes a disaggregation bias, and 3) recovery of true parameters requires assumption of the underlying advertising process. In contrast, we show that 1) the optimal data interval is what we call the unit exposure time, 2) too disaggregate data does not cause any disaggregation bias, and 3) recovery of true parameters does not require an assumption of the advertising process but only data at the unit exposure time. These results hold for any linear dynamic model linking sales with current and past advertising.

Abstract:
The popular 'airline' model for a seasonal time series assumes that a variable needs double differencing, i.e. first and seasonal (or annual) differencing. The resultant time series can usaually be described by a low order moving average model with estimated roots close to the unit circle. This latter feature complicates the standard autoregression-based tests for (seasonal) unit roots which are often used in practice. In this paper we propose an alternative route to detect seasonal unit roots by analysing (version of) the basic structual model [BSM].

Abstract:
A key application of long memory time series models concerns inflation. Long memory implies that shocks have a long-lasting effect. It may however be that empirical evidence for long memory is caused by neglecting one or more level shifts. Since such level shifts are not unlikely for inflation, where the shifts may be caused by sudden oil price shocks, we examine whether evidence for long memory (indicated by the relevance of an ARFIMA model) in G7 inflation rates is spurious or exaggerated. Our main findings are that apparent long memory is quite resistant to level shifts, although for a few inflation rates we find that evidence for long memory disappears.

Abstract:
In this paper we propose a model selection strategy for a univariate periodic autoregressive time series which involves tests for one or more unit roots and for parameter restrictions corresponding to seasonal unit roots and multiple unit roots at the zero frequency. Examples of models that are considered are variants of the seasonal unit roots model and the periodic integration model. We show that the asymptotic distributions of various statistics are the same as well-known distributions which are already tabulated. We apply our strategy to three empirical series to illustrate its ease of use. We find that evidence for seasonal unit roots based on nonperiodic models disappears when periodic representations are considered.

Abstract:
Unit root tests and cointegration tests are sensitive to atypical events as outliers and structural breaks. This paper uses outlier robust estimation techniques to reduce the impact of these events on cointegration analysis. As a byproduct of computing the robust estimator, we obtain weights for all observations in the sample. These weights can be used to identify approximate dates of those atypical events. We evaluate our method via some illustrative simulated data. Furthermore, since our robust approach involves a few additional decisions on the values of key parameters, we investigate the sensitivity of our method through extensive Monte- Carlo simulations. Finally, we present an empirical example based on real-life data to show that OLS based cointegration can yield spurious cointegration.

Abstract:
In this paper we propose a simple method to measure the impact of promotional activities on weekly market share. The main idea is to assume that if promotion has an effect, it generates an additive outlier or a temporary level shift in the market share data. We propose an outlier robust estimation technique that can give estimates of the size of such an additive outlier or temporary level shift relative to an outlier-free time series. These estimated sizes then measure the impact of promotion. We illustrate our method for two examples concerning market shares of fast moving consumer products. Two recent surveys on the analysis of the effect of promotional activities on sales and market share in Blattberg and Neslin (1989) and Blattberg, Briesch and Fox (1995) conclude with many interesting questions for further research. One of these involves the design of proper econometric methods to examine static and/or dynamic effects of promotion. In the present paper we aim to contribute to this important research area by proposing a simple econometric time series technique (based on robust estimation methods) that can estimate the net effect of promotion from noisy data. The main idea of our approach is that we assume that promotional activities generate outliers or level shifts in the market share data. We apply our technique to more than two years of weekly scanning data of the market shares of two brands of a fast-moving consumer product. A useful advantage of our approach is that we are able to estimate the so-called "baseline" market share at the time promotion occurred (see Blattberg and Neslin, 1989, p. 89) and also that we can provide confidence intervals for the quantitative effect of promotion. An alternative to our methodology would be to use zero-one dummy variables in a time series regression (see Leone, 1987, for a marketing application of such so-called intervention analysis). Application of our robust technique, however, relieves the practitioner from the burdensome task of specifying the correct delay effects of promotional activities on market share, something which cannot be avoided when using the dummy approach.

Abstract:
In the present paper we confine ourselves to proposing tests for smooth transition nonlinearity in the presence ou outliers. We consider outlier robust estimation techniques to modify the tests developed by Luukkonen et al.

Abstract:
Changing seasonal patterns in economic time series can be described by auregressive models with seasonal unit roots or with deterministic sesaonal mean shifts.By means of simulation we demonstrate the impact of imposing the incorrect model on forecasting. We find for both cases that an inappropriate decision can deteriorate forecasting performance dramatically.

Abstract:
In this paper we consider model selection for time series with increasing (or decreasing) seasonal variation, where this variation can be described by (seasonal) unit root models with significant deterministic components or by models with less unit roots but with shiftsin seasonal means or trends.

Abstract:
In this paper we show that the answer to the question in the title is affirmative, i.e. seasonal adjustment increases the probabilities in a Markov switching regime model of staying in the same regime. This phenomenon is illustrated through Monte Carlo Simulations and with two examples concerning German unemployment and US industrial production.

Abstract:
In this paper we show that the answer to the question in the title is affirmative, i.e. seasonal adjustment increases the probabilities in a Markov switching regime model of staying in the same regime. This phenomenon is illustrated through Monte Carlo Simulations and with two examples concerning German unemployment and US industrial production.

Abstract:
Standard unit root tests and cointegration tests are sensitive to atypical events such as outliers and structural breaks. This paper uses outlier robust estimation techniques to reduce the impact of these events on cointegration analysis. As a byproduct of computing the robust estimator, we obtain weights for all observations in the sample. These weights can be used to identify the approximate dates of the atypical events. We evaluate our method using some illustrative simulated data. Furthermore, since our robust approach involves a few additional decisions on the values of key parameters, we investigate the sensitivity of our method through extensive Monte-Carlo simulations. Finally, we present an empirical example based on real-life data to show that OLS-based cointegration tests can spuriously indicate stationarity.

Abstract:
In this paper we investigate the properties of the Lagrange Multiplier (LM) test for autoregressive conditional heteroskedasticity (ARCH) and generalized ARCH (GARCH) in the presence of additive outliers (AO's). We show analytically that both the asymptotic size and power are adversely affected if AO's are neglected: the test rejects the null hypothesis of homoskedasticity too often when it is in fact true, while the test has difficulty detecting genuine GARCH effects. Several Monte Carlo experiments show that these phenomena occur in small samples as well. We design and implement a robust test, which has better size and power properties than the conventional test in the presence of AO's. Applications to the French industrial production series and weekly returns of the Spanish peseta/US dollar exchange rate reveal that, sometimes, apparent GARCH effects may be due to only a small number of outliers and, conversely, that genuine GARCH effects can be masked by outliers.

Abstract:
In this paper we consider empirical econometric models for nine brands of a fast-moving nondurable consumer product using weekly observed scanning data on market share and distribution conditional on advertising, price, and promotion activities. Since the data show nonstationary characteristics, we rely on cointegration techniques to estimate long-run and short-run parameters. Additionally, as there are many outlying observations in our weekly scanning data, we apply robust cointegration methods. We find different results across robust and non-robust methods for the long-run relations between market share and distribution and for the short-run response to disequilibrium situations.

Abstract:
In this paper we test for (Generalized) AutoRegressive Conditional Heteroskedasticity [(G)ARCH] in daily and weekly data on 22 exchange rates and 13 stock market indices using the standard Lagrange Multiplier [LM] test for GARCH and a new LM test that is resistant to additive outliers. The data span two samples of 5 years ranging from 1986 to 1995. Our main result is that we find spurious GARCH in over 50% of the cases. Using Monte Carlo simulations, in which we evaluate our empirical method, we show that this general finding indeed appears to be due to outliers. We discuss some of the implications of our findings for empirical financial modeling.