Corporate insider trading as a stock market and economic indicator
Insider trading has been used in stock market forecasting for several decades. This study introduced new approaches to the research of corporate insider trading and its potential in stock market and economic forecasting. This is the first research based on the aggregate insider trading data. This is also the most comprehensive study to date comparing the performance of indicators based on the number of transactions, the value of transactions and the number of shares traded to date. In general, the study shows that insider trading contains very useful information regarding future performance of the economy and of the stock market. However, this information is contaminated with a lot of noise. Actually, the degree of contamination is such that indicators developed on the basis of unrefined aggregate insider trading shows no statistically significant relationship to the stock market or to economic variables. Therefore, the critical element in the creation of useful insider trading indicators is the application of various filters, i.e., caps on transaction values or elimination of multiple transactions of the same insiders combined with selection according to both unanimity and intensity principles. The study has shown that value-based indicators are superior to the indicators obtained with the number of transactions and the number of shares traded. This is particularly true when insider trading has been combined with macroeconomic variables and dummy variables, which describe relative over- and undervaluation of the stock market, in the creation of successful models predicting stock market movements. These broad models are capable of explaining over seventy percent of the total variation in the S&P 500 index. The buy-to-sell ratios commonly used as the measure of insiders' sentiment regarding future stock market performance, was found to have no statistically significant relationship to future levels of the stock market. This research showed that insider trading can be used in the forecasting of economic variables. Again, dollar-based indicators, with only one exception, proved to be superior indicators as compared to indicators calculated with the number of transactions and the number of shares traded.
Dadak, Kazimierz, "Corporate insider trading as a stock market and economic indicator" (1992). ETD Collection for Fordham University. AAI9223810.