• Selected
• Entire List

Abstract:
A common approach to estimating the conditional volatility of short horizon asset returns is to use an exponentially weighted moving average (EWMA) of squared past returns. The EWMA estimator is based on the maximum likelihood estimator of the variance of the normal distribution, and is thus optimal when returns are conditionally normal. However, there is ample evidence that the conditional distribution of short horizon financial asset returns is leptokurtic, and so the EWMA estimator will generally be inefficient in the sense that it will attach too much weight to extreme returns. In this paper, we propose an alternative EWMA estimator that is robust to leptokurtosis in the conditional distribution of portfolio returns. The estimator is based on the maximum likelihood estimator of the standard deviation of the Laplace distribution, and is a function of an exponentially weighted moving average of the absolute value of past returns, rather than their squares. We employ the robust EWMA estimator to forecast the VaR of aggregate equity portfolios for the US, the UK and Japan using historical simulation. We find that the robust EWMA estimator offers an improvement over the standard EWMA estimator. In particular, the VaR forecasts that it generates are as accurate as those generated by the standard EWMA estimator, but are more efficient in the sense that the average level of capital required to cover against unexpected losses is lower and the root mean square deviation between the VaR forecast and actual returns is smaller. Moreover, the volatility of the VaR forecast itself is substantially lower with the robust EWMA estimator than with the standard EWMA estimator, reflecting its lower sensitivity to extreme returns.

Abstract:
In this paper, we show that although minimum-variance hedging unambiguously reduces the standard deviation of portfolio returns, it tends to increase portfolio kurtosis and consequently the effectiveness of hedging in terms of a more general measure of risk such as VaR is uncertain. We compare the reduction in standard deviation with the reduction in 99% VaR for thirteen cross-hedged currency portfolios using both in-sample and out-of-sample approaches. We find that minimum-variance hedging reduces standard deviation considerably more than it reduces VaR. Indeed, for some portfolios, the out-of-sample reduction in VaR is negligible. As an alternative, we propose a minimum-VaR hedging strategy that minimises the historical simulation VaR of the hedge portfolio. Minimum-VaR hedge ratios are found to be significantly lower than minimum-variance hedge ratios. The minimum-VaR hedging strategy offers a significant improvement over the minimum-variance hedging strategy in terms of VaR. Moreover, in many cases, it actually yields a larger out-of-sample reduction in standard deviation also.

Hedging, Value at risk, Skewness, Kurtosis, Historical simulation

Abstract:
In this paper, we develop a momentum trading strategy based on the low frequency trend component of the spot exchange rate. Using, alternately, kernel regression and the high-pass filter of Hodrick and Prescott (1997), we recover the non-linear trend in the monthly exchange rate and use short-term momentum in this to generate buy and sell signals. The low frequency momentum trading strategy offers greater directional accuracy, higher returns and Sharpe ratios and lower maximum drawdown than traditional moving average rules. Moreover, unlike traditional moving average rules, the performance of the low frequency momentum trading strategy is relatively robust across different time periods, and to the choice of smoothing parameters across a wide range of values.

Momentum, Moving average rules, Hodrick-Prescott filter, Kernel regression, Trading strategy

Abstract:
There has recently been renewed interest in the intraday range (defined as the difference between the intraday high and low prices) as a measure of local volatility. Recent studies have shown that estimates of volatility based on the range are significantly more efficient than estimates based on the daily close-to-close return, are relatively robust to market microstructure noise, and are approximately log-normally distributed. However, little attention has so far been paid to forecasting volatility using the daily range. This is partly because there exists no multivariate analogue of the range and so its use is limited to the univariate case. In this paper, we propose a simple estimator of the multivariate conditional variance-covariance matrix of returns that combines both the return-based and range-based measures of volatility. The new estimator offers a significant improvement over the equivalent return-based estimator, both statistically and economically.

Conditional variance - covariance matrix of returns, Exponentially weighted moving average (EWMA), Intraday range

Abstract:
This paper tests the expectations hypothesis of the term structure using cross-section bond yield data. A long series of monthly cross-section regressions is estimated using zero coupon bond yields for maturities from two months to thirty-five years. The expectations hypothesis is tested using the time-series average of the estimated slope parameter in the cross-section regressions. To allow for maturity-specific, and possibly time-varying risk, the squared excess holding period return for each bond is used as a proxy for the risk premium and the regressions are estimated by instrumental variables. In strong contrast with existing evidence from time-series tests, it is found that the expectations hypothesis cannot be rejected, with estimated parameters close to their hypothesised values. The risk premium itself is significant only for the sub-sample of short maturity bonds.

Abstract:
The expectations hypothesis of the term structure of interest rates implies that the spread between short and long bond yields should forecast next period's change in the long yield. Regression based tests have systematically rejected the expectations hypothesis, with estimated coefficients far from their hypothesised values. One explanation of this rejection is that regression tests fail to account for time varying risk premia that are correlated with the spread, causing a downward bias in the estimated regression parameters. This paper uses panel data in order to test the expectations hypothesis in the presence of time varying risk premia. It is assumed that risk premia are driven by single factor, and that they are linear in bond maturity. This allows the unobserved time varying risk premia to be captured by time-specific fixed effects in a panel data regression. The hypothesis that risk premia are linear in maturity is tested by the inclusion of bond specific fixed effects in the regression. The bond specific effects are not significant, implying that risk premia are well approximated by a linear single factor model. When risk premia are handled in this way, the estimated coefficient in the regression remains significantly less than unity but is insignificantly different from zero, implying that in contrast with existing studies yield spreads have no predictive power for changes in the long yield.

Abstract:
This paper proposes a simplified multivariate GARCH model that involves the estimation of only univariate GARCH models, both for the individual return series and for the sum and difference of each pair of series. The covariance between each pair of return series is then imputed from these variance estimates. The model that we propose is considerably easier to estimate than existing multivariate GARCH models and does not suffer from the convergence problems that characterize many of these models. Moreover, the model can be easily extended to include more complex dynamics or alternative forms of the GARCH specification. We use the simplified multivariate GARCH model to estimate the minimum-variance hedge ratio for the FTSE 100 index portfolio, hedged using index futures, and compare it to four of the most widely used multivariate GARCH models. The simplified multivariate GARCH model performs at least as well as the other models that we consider, and in some cases better than them.

Multivariate GARCH, hedging, minimum-variance hedge ratio, FTSE 100 index

Abstract:
In this paper, we document a very strong day-of-the-month effect in the performance of momentum strategies in the foreign exchange market. We show that this seasonality in trading strategy performance is attributable to seasonality in the conditional volatility of foreign exchange returns, and in the volatility of conditional volatility. Indeed a two-factor model employing conditional volatility and the volatility of conditional volatility explains as much as 70 percent of the intra-month variation in the Sharpe ratio. We further show that the seasonality in volatility is in turn closely linked to the pattern of US macroeconomic news announcements, which tend to be clustered around certain days of the month.

Momentum, Moving average rules, Seasonality, Conditional volatility

Abstract:
There is now widespread evidence that investment strategies based on value and momentum have been profitable in the past. Moreover, combining value and momentum into a single investment strategy provides investment performance that is less sensitive to market cyclicality. However, portfolio managers may be reluctant to implement such strategies as they can lead to substantial departures from client-assigned benchmarks. In this paper, we implement a combined value-momentum strategy using the Black-Litterman portfolio optimisation framework, applied to a single global market comprising 177 national industry indices of the US, UK and Japan. We develop forecasting models for zero-investment value and momentum strategies, and incorporate the out-of-sample forecasts from these models into the Black-Litterman approach. The combined value-momentum strategy yields a significant improvement in performance relative to the underlying benchmark. Using the Black-Litterman model, we can effortlessly track the benchmark at the desired tracking error level under full investment, long-only and beta-neutral constraints, while producing an average annual investment outperformance of up to 0.7 percent, even after allowing for substantial transaction costs.

Portfolio optimisation, Value; Momentum, Black-Litterman, Trading strategies.

Abstract:
Models of the time-varying conditional minimum-variance hedge ratio (MVHR) typically do not provide a significant improvement in terms of hedging performance over the unconditional MVHR model. In view of the widely documented success of conditional volatility models (on which models of the conditional MVHR are usually based), this is somewhat surprising. In this paper, using the recently developed realized beta framework of Andersen, Bollerslev, Diebold and Wu (2005), we explore the reasons for this finding. We firstly show that the reduction in hedged portfolio variance that conditional MVHR models offer falls far short of the ex post maximal reduction in variance obtained using an estimate of the unobserved 'integrated' MVHR. We investigate the statistical properties of the forecasts of conditional MVHR models and show that while they do contain significant information about the integrated MVHR, they are systematically biased and inefficient. However, correcting for this bias and inefficiency does little to improve their hedging performance, suggesting that their poor performance is more likely to be attributable to the unpredictability of the integrated MVHR.

Minimum-variance hedge ratio, Realized beta, Multivariate conditional volatility models, Bias correction

Abstract:
Many applications in finance use a non-linear transformation of the variance of returns. While the sample variance is an unbiased and consistent estimator of the population variance of returns, non-linear transformations of the sample variance will be consistent but biased. For estimates of non-linear transformations of the unconditional variance, this will rarely be a problem in practice, since sample sizes employed in finance are typically large. However, estimators of the conditional variance typically use sample sizes that are effectively much smaller, particularly those that apply an exponential weighting to returns such as GARCH or EMWA. Consequently, the bias is likely to be more important in estimating non-linear transformations of the conditional variance. In this paper, we derive a simple analytical approximation for the unconditional bias in estimators of non-linear transformations of the conditional variance, under the assumption that returns are conditionally normally distributed, and that the true conditional variance is generated by an arbitrary stochastic volatility model. As an illustration, we estimate the bias inherent in the RiskMetrics approach to the calculation of value at risk.

Conditional variance, Conditional standard deviation, Non-linear transformation, Small sample bias, EWMA, GARCH, Value at risk

Abstract:
The non-normality of financial asset returns has important implications for hedging. In particular, in contrast with the unambiguous effect that minimum-variance hedging has on the standard deviation, it can actually increase the negative skewness and kurtosis of hedge portfolio returns. Thus the reduction in Value at Risk (VaR) and Conditional Value at Risk (CVaR) that minimum-variance hedging provides can be significantly lower than the reduction in standard deviation. In this paper, we provide a new, semi-parametric method of estimating minimum-VaR and minimum-CVaR hedge ratios based on the Cornish-Fisher expansion. We apply the method to four equity index positions hedged with equity index futures. We find that the semi-parametric approach is superior to the standard minimum-variance approach, and to the non-parametric approach of Harris and Shen (2006). In particular, it provides a greater reduction in both negative skewness and excess kurtosis, and consequently generates hedge portfolios that have lower VaR and CVaR.

Equity index futures, Hedging, Value at risk, Conditional value at risk, Skewness, Kurtosis, Cornish-Fisher expansion

Abstract:
The question of rules versus discretion has generated a great deal of debate in many areas of the social sciences. Recently, much of the discussion among academics and stakeholders about the assessment of research in UK higher education institutions has focused on the means that should be used to determine research quality. We present a model of committee decision-making when both rules and discretion are available. Some of the predictions of the model are tested empirically using the UK RAE 2001 results.

Rules, Discretion, Research Assessment Exercise, Probit

Abstract:
The expectations hypothesis of the term structure has been decisively rejected by a large empirical literature that spans several decades. In this paper, using a newly constructed dataset of synthetic zero coupon bond yields, we show that evidence against the expectations hypothesis became very much weaker following the widespread acceptance of its empirical failure to describe the behavior of interest rates in the early 1990s. Indeed, in the period 1991-2004, the expectations hypothesis cannot be rejected for most bond maturities. These results are consistent with the idea that asset pricing anomalies tend to disappear once they are widely recognized.

Expectations hypothesis of the term structure of interest rates, Forward yields, Yield spreads, Campbell and Shiller tests, Vector autoregression

Abstract:
Empirical rejections of the rational expectations hypothesis (REH) in the bond market have attracted much attention. In this paper we demonstrate that if agents have information about next period's short yield in addition to that contained in the current short yield, a small sample bias arises in conventional regression tests of the REH. We show that this bias may serve to significantly weaken the rejection of the REH that has been reported in the literature.

Rational Expectations Hypothesis, Term Structure of Interest Rates, Finite Sample Bias, Monte Carlo Simulation

Abstract:
This paper investigates return and volatility spillover effects between the FTSE 100, FTSE 250 and FTSE Small Cap equity indices using the multivariate GARCH framework. We find that return and volatility transmission mechanisms between large and small stocks in the UK are asymmetric. In particular, there are significant spillover effects in both returns and volatility from the portfolios of larger stocks to the portfolios of smaller stocks. For volatility, there is also evidence of limited feedback from the portfolios of smaller stocks to the portfolios of larger stocks, although sub-period analysis suggests that this is to some extent period-specific. Simulation evidence shows that non-synchronous trading potentially explains some, but not all, of the spillover effects in returns, and that it explains none of the spillover effects in volatility. These results are consistent with a market in which information is first incorporated into the prices of large stocks before being impounded into the prices of small stocks.

Abstract:
This paper proposes a simplified multivariate GARCH model (the S-GARCH model) that involves the estimation of only univariate GARCH models, both for the individual return series and for the sum and difference of each pair of series. The covariance between each pair of return series is then imputed from these variance estimates. The model that we propose is considerably easier to estimate than existing multivariate GARCH models and does not suffer from the convergence problems that characterize many of these models. Moreover, the model can be easily extended to include more complex dynamics or alternative forms of the GARCH specification. We use the S-GARCH model to estimate the minimum-variance hedge ratio for the FTSE 100 index portfolio, hedged using index futures, and compare it to four of the most widely used multivariate GARCH models. Using both statistical and economic evaluation criteria, we find that the S-GARCH model performs at least as well as the other models that we consider, and in some cases better than them.

Multivariate GARCH, Hedging, Minimum-variance hedge ratio, FTSE 100 index

Abstract:
A number of financial variables have been shown to be effective in explaining the time-series of aggregate equity returns in both the UK and the US. These include, inter alia, the equity dividend yield, the spread between the yields on long and short government bonds, and the lagged equity return. Recently, however, the ratio between the long government bond yield and the equity dividend yield ? the gilt-equity yield ratio ? has emerged as a variable that has considerable explanatory power for UK equity returns. This paper compares the predictive ability of the gilt-equity yield ratio with these other variables for UK and US equity returns, providing evidence on both in-sample and out-of-sample performance. For UK monthly returns, it is shown that while the dividend yield has substantial in-sample explanatory power, this is not matched by out-of sample forecast accuracy. The gilt-equity yield ratio, in contrast, performs well both in-sample and out-of-sample. Although the predictability of US monthly equity returns is much lower than for the UK, a similar result emerges, with the gilt-equity yield ratio dominating the other variables in terms of both in-sample explanatory power and out-of-sample forecast performance. The gilt-equity yield ratio is also shown to have substantial predictive ability for long horizon returns.

Abstract:
This paper proposes a similar unit root testing procedure for heterogeneous dynamic panel data, based on the score principle, assuming that the time dimension of the panel is fixed.

Abstract:
This paper evaluates analysts' consensus long run earnings growth forecasts. It is shown that the correlation between forecast earnings growth and actual earnings growth is extremely low. Consistent with other studies, forecast earnings growth is found to be too optimistic. This is illustrated by the fact that almost all earnings growth forecasts are positive, while actual earnings growth is more evenly distributed between positive and negative values. For the sample of companies for which actual earnings growth is positive, there is a strong positive correlation between forecast earnings growth and actual earnings growth. In sharp contrast, for the sample of companies for which actual earnings growth is negative, there is a strong inverse correlation between forecast earnings growth and actual earnings growth. This suggests that analysts are to some extent able to forecast the magnitude of earnings growth but not its sign. It is further shown that analysts' forecasts of long run earnings growth are incorporated into the market's expectation of future earnings growth. However, there is evidence that the market attaches more weight to forecasts that have the correct sign, even though the sign of actual earnings growth is not known at the date that the forecast is made.