• Selected
• Entire List

Abstract:
We document the existence of a strong seasonal effect in stock returns based on the popular market saying Sell in May and go away, also known as the Halloween indicator. According to these words of market wisdom, stock market returns should be higher in the November-April period than those in the May-October period. Surprisingly, we find this inherited wisdom to be true in 36 of the 37 developed and emerging markets studied in our sample. The Sell in May effect tends to be particularly strong in European countries and is robust over time. Sample evidence, for instance, shows that in the UK the effect has been noticeable since 1694. While we have examined a number of possible explanations, none of these appears to convincingly explain the puzzle.

Stock returns, Sell in May, Return predictability, Halloween indicator

Abstract:
We document the existence of a strong seasonal effect in stock returns based on the popular market saying 'Sell in May and go away', also known as the 'Halloween indicator'. According to these words of market wisdom, stock market returns should be higher in the November-April period than those in the May-October period. Surprisingly, we find this inherited wisdom to be true in 36 of the 37 developed and emerging markets studied in our sample. The 'Sell in May' effect tends to be particularly strong in European countries and is robust over time. Sample evidence, for instance, shows that in the UK the effect has been noticeable since 1694. While we have examined a number of possible explanations, none of these appears to convincingly explain the puzzle.

Stock returns, Sell in May, Return predictability, Halloween indicator

Abstract:
Changes in oil prices predict stock market returns worldwide. In our thirty year sample of monthly returns for developed stock markets, we find statistically significant predictability for twelve out of eighteen countries as well as for the world market index. Results are similar for our shorter time series of emerging markets. We find no evidence that our results can be explained by time varying risk premia. Even though oil price shocks increase risk, investors seem to underreact to information in the price of oil: a rise in oil prices does not lead to higher stock market returns, but drastically lowers returns. For instance, an oil price shock of one standard deviation (around 10 percent) predictably lowers world market returns by one percent. Oil price changes also significantly predict negative excess returns. Our findings are consistent with the hypothesis of a delayed reaction by investors to oil price changes. In line with this hypothesis the relation between monthly stock returns and lagged monthly oil price changes becomes substantially stronger once we introduce lags of several trading days between monthly stock returns and lagged monthly oil price changes.

Return Predictability, Oil Prices, International Stock Markets, Market Efficiency, Stock Returns, Underreaction

Abstract:
We find that changes in oil prices strongly predict future stock market returns in many countries in the world. In our thirty year sample of monthly data for developed stock markets, we find statistically significant predictability in 12 out of the 18 countries and in a world market index. For our shorter time series of emerging markets we obtain similar results. We show that these results are economically significant and robust with respect to the sample period, different kind of oil prices we consider and well known effects like the January effect and the Halloween effect.

return predictability, oil prices, international stock markets, market efficiency, stock returns, besliskunde

Abstract:
Changes in oil prices predict stock market returns worldwide. In our thirty year sample of monthly returns for developed stock markets, we find statistically significant predictability for twelve out of eighteen countries as well as for the world market index. Results are similar for our shorter time series of emerging markets. We find no evidence that our results can be explained by time varying risk premia. Even though oil price shocks increase risk, investors seem to underreact to information in the price of oil: a rise in oil prices does not lead to higher stock market returns, but drastically lowers returns. For instance, an oil price shock of one standard deviation (around 10 percent) predictably lowers world market returns by one percent. Oil price changes also significantly predict negative excess returns. Our findings are consistent with the hypothesis of a delayed reaction by investors to oil price changes. In line with this hypothesis the relation between monthly stock returns and lagged monthly oil price changes becomes substantially stronger once we introduce lags of several trading days between monthly stock returns and lagged monthly oil price changes.

Return Predictability, Oil Prices, International Stock Markets, Market Efficiency, Stock Returns, Underreaction

Abstract:
According to conventional market wisdom a sector rotation strategy over different stages of the business cycles outperforms the market. We introduce a simple way to test its value. In our test an investor anticipates business cycle stages perfectly and rotates sectors following popular belief. Even with perfect foresight and ignoring transactions costs sector rotation would have generated at best a 2.3% annual outperformance since 1948. In more realistic settings outperformance quickly dissipates. We do find an alternative rotation strategy that historically would have beaten the market by 7%. Whether by chance or due to fundamentals time will tell.

stock market, sector rotation, business cycles, investment strategies

Abstract:
Monthly stock market returns are predictable when we refine the observation intervals of the variables used to predict these returns. Contrary to other predictability studies we find high out-of-sample adjusted R2s of up to 7% using economically important commodity returns. Shorter intervals reveal predictability consistent with near efficient markets based on price changes in industrial metals. More historical intervals expose predictability consistent with gradual information diffusion based on energy series. This predictability is robust to data mining adjustment, the inclusion of control (including economic) variables, and unrelated to time-varying risk. Inflation explains part of this predictability, but not all.

observation interval, return predictability tests, market efficiency, gradual information diffusion, market timing, quantitative investment techniques, commodity

Abstract:
All US stock market sectors and industries perform better during winter than during summer in our sample from 1926-2005. In more than two-third of all sectors and industries this difference in summer and winter returns, known as the Halloween effect, is statistically significant and in half of all sectors and industries risk premia are negative during summer. However, while all sectors and industries show this effect, there are large differences across sectors and industries. The effect is almost absent in sectors related to consumer consumption but strong in production sectors. We illustrate how these differences between sectors might be used to improve the risk return trade off using sector rotation based on Fidelity sector funds and show how an investor might have benefited from such a trading strategy.

Sectors, Industries, Stock Markets, Halloween effect, Sell in May, Sector Rotation

Abstract:
In this paper we study the economic significance of simple time series models of stock return predictability. We investigate the practical usefulness of recent findings on time series return predictability of stock returns and their volatility for dynamic tactical asset allocation decisions. We introduce a mean variance investor with an investment horizon of one year who takes investment decisions daily. When stock returns follow a random walk this investor holds constant proportions of a stock market index and a risk free asset. Using past data and knowledge of some well known return predictability results (i.e. predictability based on calendar anomalies and predictability from economic variables like dividend yields and short term interest rates), we evaluate whether, how and to what extent these predictability results might affect his investment decisions. For this investor we also investigate the practical usefulness of knowledge about the predictability over time of market volatility. The design we choose is as follows. We give the predictability results the benefit of the doubt and assume that all the estimates and models are correct and indeed accurately describe the true return generating process. We then analyze analytically, numerically and by Monte Carlo simulation the effect of investment decisions--conditional on these predictability results--on the return distribution of his portfolio. We also introduce transactions costs in this setting; these influence investment choices. Our main findings are that small transactions costs substantially reduce potential benefits of trading on calendar anomalies. Generally, however trading remains profitable under the assumption that stock market returns are partially predictable from economic variables like dividend yields and interest rates. This holds for relatively large transactions costs. Trading on volatility predictability is not profitable, except in the case of negligible transactions costs.

Abstract:
Recent international evidence shows that in many stock markets, general index returns are significantly higher during winter months than during summer months. We study the interaction between this anomaly - known as the Halloween effect - and the January effect and other well-known anomalous findings on portfolios formed on Size, Dividend Yield, Book to Market ratios, Earnings Price ratios and Cash Flow Price ratios in equally but also value weighted portfolios for the US market. Our main findings are that contrary to the January effect, the Halloween effect seems a market wide phenomenon unrelated to these well-known anomalies. All portfolios in our study show higher average winter returns than summer returns. In most portfolios this difference is statistically and economically significant. We confirm recent results which suggest that the January effect plays an important role not only in explaining the small firm effect but also - together with size - in explaining the Book to Market ratio anomaly. In addition, we find in a similar fashion that controlling for the January effect and using value weighted portfolio returns substantially reduces the Earnings to Price, Cash Flow to Price and Dividend Yield effects.

Halloween Effect, Sell in May, January effect, Book to Market, Value, Growth

Abstract:
We show that results in the recent strand of the literature, which tries to explain stock returns by weather induced mood shifts of investors, might be data-driven inference. More specifically, we consider two recent studies (Kamstra, Kramer and Levi, 2003a and Cao and Wei, 2005) that claim that a seasonal anomaly in stock returns is caused by mood changes of investors due to lack of daylight and temperature variations, respectively. While we confirm earlier results in the literature that there is indeed a strong seasonal effect in stock returns in many countries: stock market returns tend to be significantly lower during summer and fall months than during winter and spring months as documented by Bouman and Jacobsen (2002), there is little evidence in favor of a SAD or temperature explanation. In fact, we find that a simple winter/summer dummy best describes this seasonality. Our results suggest that without any further evidence the correlation between weather-related variables and stock returns might be spurious and the conclusion that weather affects stock returns through mood changes of investors is premature.

Stock market seasonality, Sell in May, Seasonal affective disorder, Temperature effect, Spurious correlations

Abstract:
We show that results in the recent strand of the literature that tries to explain stock returns by weather induced mood shifts of investors might be data-driven inference. More specifically, we consider two recent studies (Kamstra, Kramer and Levi, 2003a and Cao and Wei, 2004) that claim that a seasonal anomaly in stock returns is caused by mood changes of investors due to lack of daylight and temperature variations, respectively. We confirm earlier results in the literature that there is indeed a strong seasonal effect in stock returns in many countries: stock market returns tend to be significantly lower during summer and fall months than during winter and spring months. However, we also show that at best, these two studies offer two of many possible explanations for the observed seasonal effect. As an illustration we link ice cream production and airline travel to the stock market seasonality using similar reasoning. Our results suggest that without any further evidence the correlation between weather variables and stock returns might be spurious and the conclusion that weather affects stock returns through mood changes of investors is premature.

Important Note: Results in this version deviate slightly from the Journal of Banking version as part of the dataset in that version was incorrect. Although results are qualitatively similar this version contains estimates and tables recalculated based on the proper dataset.

stock market seasonality, sell in May, seasonal affective disorder, temperature, spurious correlations

Abstract:
We show that results in the recent strand of the literature, which tries to explain stock returns by weather induced mood shifts of investors, might be data-driven inference. More specifically, we consider two recent studies (Kamstra, Kramer and Levi, 2003a and Cao and Wei, 2005) that claim that a seasonal anomaly in stock returns is caused by mood changes of investors due to lack of daylight and temperature variations, respectively. While we confirm earlier results in the literature that there is indeed a strong seasonal effect in stock returns in many countries: stock market returns tend to be significantly lower during summer and fall months than during winter and spring months as documented by Bouman and Jacobsen (2002), there is little evidence in favor of a SAD or temperature explanation. In fact, we find that a simple winter/summer dummy best describes this seasonality. Our results suggest that without any further evidence the correlation between weather-related variables and stock returns might be spurious and the conclusion that weather affects stock returns through mood changes of investors is premature.

Important Note: Results in this version deviate slightly from the Journal of Banking version as part of the dataset in that version was incorrect. Although results are qualitatively similar this version contains estimates and tables recalculated based on the proper dataset.

Stock market seasonality, Sell in May, Seasonal Affective Disorder, temperature, spurious correlations

Abstract:
A basic assumption in economics is that agents prefer more to less. We find that in an experimental financial market--a large scale political stock market--more than 50 percent of the traders act irrationally in the sense that they violate this fundamental assumption frequently (at least 10 percent of the time). However, the impact of this irrationality seems small, since related losses consist of less than 1 percent of total investments in this market and we find no effect on prices. We do however find systematic bias in prices: Our evidence shows that the sum of the bid prices of all traded stocks exceeds the fundamental value of the market about 10 percent of the time. We argue that another type of irrationality known as 'loss aversion' causes this upward bias.

Abstract:
We empirically analyze rational investors' optimal response to asset price bubbles. We define bubbles as a sudden acceleration of price growth beyond the growth in fundamental value given by an asset pricing model. Our new bubble detection method requires only a limited time-series of historical returns. We apply our method to US industries and find strong statistical and economic support for the riding bubbles hypothesis: when an investor detects a bubble, her optimal portfolio weight increases significantly. A dynamic riding bubble strategy that uses only real-time information earns abnormal annual returns of 3% to 8%.

bubbles, limits to arbitrage, market efficiency, structural breaks

Abstract:
Using a database of 440 international political crises over the period 1918-2002, we find that international crises reduce world market stock returns by approximately four percent per annum. Crises cause large negative stock market reactions in their first month, lower than average returns during the remaining months, and a partial recovery when they end. International crises not only have a strong impact on mean returns but also on volatility. The start of an average crisis increases monthly world market volatility by more than a third and the end of a crisis decreases volatility by slightly less than a third. Daily US stock market returns confirm our results. We also find that stock market reactions and volatility changes are significantly stronger when an international crisis starts with violence, involves more severe value threats, or when a major power is involved on both sides of a conflict. Financial consequences for investors in crisis actor countries are even more devastating.

War, Peace, Volatility Puzzle, Stock Returns, International Crises

Abstract:
Men are strikingly more optimistic about the future performance of key economic and financial indicators than women. We report surprisingly strong and highly significant gender differences in consumer confidence data of seventeen out of eighteen countries, including the US. We confirm these findings using data from US and European Gallup opinion polls. This gender difference is present in key indicators such as economic growth, interest rates, inflation and future stock market performance and persists after we control for income, employment, wealth, education and marital status. Our results hold regardless whether we consider questions about respondent's personal future economic situation or the general state of the economy. This suggests that the optimism we document in this study is different from the well-documented overconfidence phenomenon, as it extends beyond the personal influence sphere of respondents. We also document significant gender differences in perceived stock market risk. Thus, we show that not only differences in risk aversion but also differences in optimism and perceived risk may explain why women hold on average less risky portfolios than men.

Optimism, Pessimism, Gender Difference, Consumer Confidence, Economic Indicators, Risk Aversion

Abstract:
We find no evidence of a Daylight Saving Time anomaly in stock returns based on empirical evidence from twenty-two stock markets around the world. Stock market returns on the days following a switch from or to Daylight Saving Time do not behave any differently from stock returns on any other day of the week or month. These results reject earlier conclusions in the literature -- based on far less data -- that investors' mood changes induced by changes in sleep patterns significantly affect stock returns

Daylight Saving Time, Mood, International Stock Market Returns

Abstract:
After correcting industry returns for general market movements and known risk factors, using either the Single-Index or the Fama-French three factor models, we find no evidence of two well known political effects documented for general stock market returns in the United States. Contrary to the general market, adjusted industry returns do not show a significant or consistent underperformance under Republican presidents. Contrary to general market indices, adjusted industry returns do not exhibit significant or a consistent presidential election cycle effect. Our results defy popular belief that some industries perform consistently better under either Democrats or Republicans, and suggest these two political effects are market wide phenomena whose explanation should be sought at a macro economic level.

efficient markets, industry returns, political cycles

Abstract:
Kamstra, Kramer and Levi (KKL) in their comment seem to miss the main point of our paper. Many things are correlated with the seasons so it is difficult to distinguish between them when we try to explain the well-known summer winter pattern in stock returns. Finding an isolated Seasonal Affective Disorder (SAD) effect without proper control variables does not disprove our point but strengthens it. To sidestep all of the issues they raise and take our point to the extreme, we show using plain vanilla regressions that the seasonal stock market pattern they attribute to SAD can also be "explained" by variables like ice cream consumption or airline travel. The new variations of SAD variables ("onset" and "incidence") KKL propose in their recent work for North America are even more problematic than the original SAD variables. We find that these new SAD proxies are not significant in countries where according to KKL they should be: Canada and the United States.

Stock market seasonality, Sell in May, Seasonal affective disorder

Abstract:
U.S. stock market sectors and industries perform better during winter than summer from 1926 to 2006. In more than two-thirds of sectors and industries, the difference in summer and winter returns, known as the Halloween effect, is statistically significant. There are, however, large differences across sectors and industries. The effect is almost absent in sectors related to consumer consumption but is strong in production sectors. We find that neither liquidity changes nor well-known risk factors can explain the anomaly. We illustrate how the differences between sectors and industries can improve the risk-return tradeoff using sector rotation.

Abstract:
After correcting industry returns for general market movements and known risk factors, using either the Single-Index or the Fama-French three-factor model, we find no evidence of two well known political effects documented for general stock market returns in the United States. Contrary to the general market, adjusted industry returns do not show a significant or consistent underperformance under Republican presidents. Contrary to general market indices, adjusted industry returns do not exhibit significant or a consistent presidential election cycle effect. Our results defy popular belief that some industries perform consistently better under either Democrats or Republicans, and suggest these two political effects are market wide phenomena whose explanation should be sought at a macro economic level.

Market efficiency, Industry returns, Presidential cycle, Political business cycle

Abstract:
We show, using the modified rescaled range statistic, that none of the return series of indices of five European countries, the United States and Japan exhibits long term dependence. This statistic — introduced by Lo (1991)- correct Hurst's (1951) 'classical' rescaled range statistic for short term dependence. We also report the classical rescaled range statistic after adjusting the series for short term dependence. This procedure shows, for cases where the results of the modified rescaled range statistic are mixed, that no long term dependence can be found. Simulations indicate reasonable power of this adjustment procedure. Furthermore, we find that estimates of the Hurst exponent, a related measure of long term dependence, are also biased by short term dependence. Simulations show that this measure - that has recently attracted growing interest - cannot distinguish between models with or without long term dependence.

Time series, Stock returns, Long term dependence, Rescaled range statistic, R/S analysis, Hurst exponent

Abstract:
We investigate volatility clustering using a modeling approach based on the temporal aggregation results for generalized autoregressive conditional heteroscedasticity (GARCH) models in Drost and Nijman [Econometrica, 1993]. Our findings highlight that volatility clustering, contrary to widespread belief, is not only present in high-frequency financial data. Monthly data also exhibit significant serial dependence in the second moments. We show that the use of temporal aggregation to estimate low-frequency models reduces parameter uncertainty substantially.

Stock returns, GARCH, Volatility clustering, Temporal aggregation

Abstract:
This paper presents the results of a political stock market in the Netherlands: PAM94. The exchange covered three consecutive elections, allowing trade on five different markets. The predictions at PAM94 appear to be less accurate than those of previous markets of comparable size. Of the possible explanations that we examine, one in particular survives closer scrutiny. It concerns a type of judgement failure related to the winner's curse in common value auctions. Theoretical as well as empirical support is offered. Apart from qualifying the attractiveness of such markets as an alternative for opinion polls, this explanation may also be relevant for the analysis of other asset markets. Moreover, this judgement failure may be more important for European political stock markets than for the U.S., because the structure of the common values (vote shares) at multiparty elections make them especially vulnerable to it.

Political stock market, Common value, Proportional representation, Market efficiency

Abstract:
Stock market crashes are rare events. This complicates a thorough quantitative empirical analysis of crashes and their probable causes. We introduce the concept of an industry crash and study the presence of these crashes in 48 US industry indexes over the period 1926 to 2004. The concept of an industry crash enlarges the sample of crashes available for study substantially. This large sample of crashes allows us to test several theoretical hypotheses recently put forward in the literature on the relation between bubbles and crashes. Our empirical evidence shows that bubbles - periods of strong outperformance - double the likelihood of a crash. This relation is stronger for more severe crashes: for the twenty percent largest industry crashes the presence of a bubble triples the crash probability. Our results also confirm theoretical results that crashes are more likely when bubble growth is stronger, but that the time a bubble takes to develop seems unrelated to crash likelihood. Last but not least, we verify whether these results for industry crashes carry over to general stock market crashes using shorter time series of stock market indexes for 49 countries starting in 1969. We find that they do.

bubbles, crashes, skewness

Abstract:
We show that investigation of the return series of indices of five European countries, the United States and Japan rejects the conclusion of long term dependence in these series. The statistic used to establish this result is the 'modified' rescaled range, suggested by Lo (1991). This statistic adjusts the 'classical' rescaled range (introduced by Hurst (1951)) for short term dependence. We also report additional results of a Monte Carlo simulation to determine the empirical size and finite sample distribution of these statistics when the data exhibit so-called volatility clustering. Our simulation evidence indicates that in the presence of short term dependence and volatility clustering the modified and classical rescaled range reject the null hypothesis of no long term dependence too frequently.