Intradaily and weekly patterns in information arrival in the S&P 500 Index futures market
Abstract
This study examines whether intradaily and weekly patterns occur in three variables calculated from 1983-1988 S&P 500 Index futures transaction prices, and whether any patterns are consistent with patterns in information arrival or information trading. Intradaily and weekly patterns have been previously documented for returns, returns variance, volume, and other variables calculated from NYSE data. These patterns are usually U-shaped, e.g. index returns variance is high near the open and close of trading. Underlying patterns in information arrival and information trading have been suggested as the cause of the patterns by Jain and Joh (1988) and others. But the patterns may be related to non-synchronous trading, or to exchange-specific factors. The short time periods that have been studied limit our confidence in the generality of the patterns. A six year series of transaction data for a single asset that is not traded on the NYSE, S&P 500 Index futures, is used to examine the questions raised by these patterns. U-shaped intradaily patterns in the number of trades and price volatility are expected if a pattern in information arrival exists. But a U-shaped pattern in the proxy for the autocorrelation of transaction price changes would be inconsistent with Jain and Joh's conjecture of concentrations of information traders near the open and liquidity traders near the close. An S-shaped intradaily pattern occurs in the autocorrelation proxy in the S&P futures market. Rough U-shaped intradaily patterns are apparent in the number of trades and in price volatility. The intradaily patterns repeat for each year for each variable. The patterns change slightly across weekdays in the afternoon, resulting in U-shaped or inverted U-shaped weekly patterns. A distinctive change in the patterns occurs on Friday afternoon for all three variables. These results are, in general, supported by the results from a five year series of Eurodollar futures transaction data. These patterns are consistent with underlying patterns in information arrival and information trading.
Degree
Ph.D.
Advisors
McConnell, Purdue University.
Subject Area
Finance
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.