Theodore E. Day

Stock market volatility and the information content of stock index options

Previous studies of the information content of the implied volatilities from the prices of call options have used a cross-sectional regression approach. This paper compares the information content of the implied volatilities from call options on the S&P 100 index to GARCH (Generalized Autoregressive Conditional Heteroscedasticity) and Exponential GARCH models of conditional volatility. By adding the implied volatility to GARCH and EGARCH models as an exogenous variable, the within-sample incremental information content of implied volatilities can be examined using a likelihood ratio test of several nested models for conditional volatility. The out-of-sample predictive content of these models is also examined by regressing ex post volatility on the implied volatilities and the forecasts from GARCH and EGARCH models.

Following the leader: ☆: a study of individual analysts’ earnings forecasts

Abstract This paper develops and tests procedures for ranking the performance of security analysts based on the timeliness of their earnings forecasts, the abnormal trading volume associated with these forecasts, and forecast accuracy. Our framework provides an objective assessment of analyst quality that differs from the standard approach, which uses survey evidence to rate analysts. We find that lead analysts identified by our measure of forecast timeliness have a greater impact on stock prices than follower analysts. Further, we find that performance rankings based on forecast timeliness are more informative than rankings based on abnormal trading volume and forecast accuracy. We also present evidence that analysts forecast revisions are correlated with recent stock price performance, suggesting that security analysts use publicly available information to revise their earnings forecasts.

The behavior of the volatility implicit in the prices of stock index options

Abstract We examine stock-market volatility around the quarterly expirations of stock index futures contracts and nonquarterly expirations of stock index options, using estimates of the volatility implicit in the option prices. The option prices reflect increases in the volatility of the underlying stock indexes around both quarterly and nonquarterly expiration dates. Analysis of the residual returns on index options provides evidence consistent with an unexpected increase in market volatility around expiration dates.

Forecasting Futures Market Volatility

This article compares the accuracy o f dgerent methods o f forecasting the volatility of crude oil futures prices. Using daily datafrom November 1986 through March 1991, we examine volatilities j o m models of the generalized autoregressive conditional heteroscedasticity ( G A R C H ) family and contrast them with the implied volatilities j o m call options on crude oil futures. In-sample tests are conducted by embedding the implied volatility in GARCH and EGARCH models as an additional explanatory variable. The results show that both sources o f volatility infrmation contribute statistically signgcant mplanatory power. We j n d no particular reason to prefer the more complex EGARCH model, because there appears to be no asymmetry in the volatility response to firtures price changes. I n out-of-sample tests of forecasts of futures volatility over the remaining lives of both nearby and more distant option contracts, the GARCH and E G A R C H models violate the requirements f o r forecast rationality, but implied volatilities do not. A naive historical volatility estimate is also examined, but its perfarmance is not as good as implied volatility. “Encompassing regression” tests show that out-of-sample forecasts from the GARCH-type models contained no information that was not impounded in implied volatilities. Bias-adjusted and combined forecasts based on the encompassing regression results do not perform as well out-of-sample as the unadjusted implied volatility.

Dividends, nonsynchronous prices, and the returns from trading the Dow Jones Industrial Average

Abstract The informational efficiency of the Dow Jones Industrial Average is of particular interest given the role of the index in tracking movements in the stock market. Brock et al. [Journal of Finance 47 (1992) 1731.] found that technical rules can be used to profitably trade the Dow Jones Industrial Average, suggesting that the prices of Dow components fail to fully reflect the information in past prices. We reexamine these findings by adjusting the daily returns from trading portfolios of Dow stocks for both dividends and the interest earned on the proceeds from short sales. Our results suggest that previous estimates of trading profits may be biased by the inclusion of nonsynchronous prices in the closing index levels. We show that estimates of trading profits based on the true closing levels of the index are not significantly different from buy and hold returns. Further, we present bootstrap simulation results which suggest that significance levels for estimates of trading profits based on closing index levels are overstated by the inclusion of nonsynchronous prices.

Transfer Pricing, Incentive Compensation and Tax Avoidance in a Multi-division Firm

This article examines the relation between transfer pricing and production incentives using a model of a vertically integrated firm with divisions located in different tax jurisdictions. We show that if divisional profits are taxed at the same marginal rate, the transfer price should be set to minimize the compensation risk faced by the manager of the buying division. For the case where divisional profits are taxed at different marginal rates, we are able to characterize the trade-off between the tax savings from setting transfer prices to reduce profitability in the high tax jurisdication and the loss of effort attributable to the impact of tax avoidance on the incentive compensation system. Further, we show that if it is feasible to compensate the division managers using multiple performance measures, the transfer price should be used to minimize the firms overall tax liability. Finally, we show that when authority to determine the transfer price must be delegated to one of the division managers, it is optimal to assign responsibility for setting the transfer price to the manager of the division with the most production uncertainty.