Blog - Investing Notes

February 24, 2005 – Information-Based Trading

Anyone attempting or contemplating day-trading should take a look at "Information-based Trading, Price Impact of Trades, and Trade Autocorrelation" of May 2004 by Kee Chung, Mingsheng Li and Thomas McInish. The authors examine how the price impact of trades and the persistence of trading direction relate to the probability of information-based trading, as opposed to liquidity trading (for example, by market makers) or noise trading (by the uninformed). Here’s what they find:

• The price impacts of stock trades are positively and significantly related to the probability that the trades are based on new information about the stock.
• The higher the probability that trading is information-based, the more likely the stock is to trade persistently in the same direction (up for positive information or down for negative information). In other words, given that informed traders are trading now, they are likely to continue trading.
• Trading in small companies and trading in low-priced stocks are more likely to be information-based than other types of trading.
• Shorter intervals between trades induce larger price movements and stronger persistence in trade direction, especially when trading is information-based.

For background on measuring the probability that trading is information based, see "One Day in the Life of a Very Common Stock" by David Easley, Nicholas Kiefer and Maureen O’Hara from the Fall 1997 edition of The Review of Financial Studies.

In summary, the higher the probability that trading in a stock is informed, the greater the size and persistence of stock price movement.

See Blog Synthesis: Some Trading Indicators for analyses of the usefulness of other technical indicators. See also our blog entry of 10/22/04 for research on whether day traders make money.