Online Program

Wavelet Analysis and Directors' Dealings (308441)

Keywords: New statistical methods, Wavelet Analysis, Capital Markets, Factor Models, Insiders' trades, capital market regulation, empowering outsiders, equality on capital markets

The validity of efficient market hypotheses have been tested in a multitude of econometric settings. Directors are required to notify regulatory authorities of their dealings in companies where they are members of management and/or supervisory boards. In this paper we investigate by event studies if insiders have superior information with which the market can be outperformed in a statistically significant way. For this purpose we analyze data on insider trades from various European countries that have not been analyzed before and estimate returns on performances by insiders and outsiders. A common problem with this approach, however, is that expected returns have to be modeled using factor models like CAPM or APT which estimations can result in biases. We use modified discrete wavelet transforms to distinguish between expected returns and noise and thereby avoid those biases. The advantage of wavelet analysis is that the components can be interpreted on a time and frequency scale and are therefore explainable and not a black box. A second application for wavelet analysis in this paper is concerned with the time period insider information might be able to generate out performances.