A copula-based stochastic frontier model and efficiency analysis: Evidence from stock exchange of Thailand

Abstract

© Springer International Publishing AG 2016. This study applies the concept of stochastic frontier model (SFM) to estimate stock efficiencies of the top 50 companies with the highest market capitalization in the Stock Exchange of Thailand (SET50). We decompose the actual return deviation from its expected return into a stochastic noise and inefficiency term, and use copula approach to join these two error components. Four copulas are considered, and the most appropriate copula is selected using the lowest AIC. The empirical results show that the majority of average return efficiencies (60% of stocks) lie between 0.9571 and 0.9999, suggesting that most of the stocks are quite efficient. However, the overall average of all return efficiencies is found to be 0.7313, indicating that stock price does not reflect all relevant information in the market.