Investor confidence as a determinant in the Mexican stock market through a TAR-TGARCH model

Abstract

In this work, a two regimes TAR-EGARCH model is applied in order to study the effects of psychological biases on the capital market. Implicit volatility is included in each regimen as an indicator of fear among informed investors. Conditional variance equation includes factors that represent investor’s overconfidence to determine if this emotional bias affects returns volatility.
The results show that overconfidence is a determinant of volatility; and the regimes in the conditional mean, are determined by confidence that uninformed investors have of the economy. The fear of rational investors affects the stock returns in each regime.

JEL Classification: C22, G12, G41