This paper examines the economic value of overnight information to users of risk management models. In addition to the information revealed by overseas markets that trade during the (domestic) overnight period, this paper exploits information generated via recent innovations in the structure of financial markets. In particular, certain securities (and associated derivative products) can now be traded at any time over a 24-h period. As such, it is now possible to make use of information generated by trading, in (almost) identical securities, during the overnight period. Of the securities that are available over such time periods, S&P 500 related products are by far the most actively traded and are, therefore, the subject of this paper. Using a variety of conditional volatility models that allow time-dependent information flow within (and across) three different S&P 500 markets, the results show that overnight information flow has a significant impact on the conditional volatility of daytime traded S&P 500 securities. Moreover (time-consistent) forecasts from models that incorporate overnight information are shown to have economic value to risk managers. In particular, Value-at-Risk (VaR) models based on these conditional volatility models are shown to be more accurate than VaR models that ignore overnight information.

The objective of this paper is to uncover the determinants of trading intensity in futures markets. In particular, the time between adjacent transactions (referred to as transaction duration) on the FTSE 100 index futures market is modeled using various augmentations of the basic autoregressive conditional duration (ACD) model introduced by Engle and Russell [Econometrica 66 (1998) 1127]. The definition of transaction duration used in this paper is an important variable as it represents the inverse of instantaneous conditional return volatility. As such, this paper can also be viewed as an investigation into the determinants of (the inverse of) instantaneous conditional return volatility. The estimated parameters from various ACD models form the basis of the hypothesis tests carried out in the paper. As predicted by various market microstructure theories, we find that bid–ask spread and transaction volume have a significant impact upon subsequent trading intensity. However, the major innovation of this paper is the finding that large (small) differences between the market price and the theoretical price of the futures contract (referred to as pricing error) lead to high (low) levels of trading intensity in the subsequent period. Moreover, the functional dependence between pricing error and transaction duration appears to be non-linear in nature. Such dependence is implied by the presence of arbitragers facing non-zero transaction costs. Finally, a comparison of the forecasting ability of the various estimated models shows that a threshold ACD model provides the best out-of-sample performance.

This paper measures the economic and statistical significance of econometric forecasts of bid–ask spreads. The economic importance of these forecasts is assessed by considering the benefits of scheduling trades based on these forecasts. The unrestricted vector autoregression (VAR) model of Huang and Masulis [Rev. Financial Studies 12 (1999) 61] and the two-equation structural model of Huang and Stoll [Rev. Financial Studies 7 (1994) 179] are used to generate intraday h-step ahead forecasts of spreads for 50 stocks listed on the London Stock Exchange (LSE). The period corresponding to the minimum expected spread is then scheduled into the trading activity of the investor. The results indicate that when the unrestricted VAR model is used, the spreads incurred are around 35% lower than the spreads incurred by investors who do not schedule their trades. By contrast, spread discounts of only 5% are obtained when the two-equation structural model is used. The heterogeneity of the economic importance of the spread forecasts generated by the models is confirmed by tests of the statistical significance of the forecasts.