Portfolio selection problem based on exponential loss function under independent binomial model in SET50

Abstract

In this thesis, we investigate the portfolio allocation problem in the Stock Exchange of Thailand by considering the stocks of the 48 biggest companies in terms of market capitalization, with the requirement that an investor wants to invest under these 48 assets only and their risk is measured by the exponential loss function. The stocks under consideration must be independent and the prices of the stocks at the maturity time are assumed to follow the well-known binomial model. The Pearson correlation test is used for selecting stocks with independence property. In this study, we are interested in an investor investing in two, three, and four distinct stocks. Anywise, we suppose that an investor buys assets at an initial time with an initial wealth and holds them until maturity time without rebalancing. Finally, we examine the efficiency of the obtained optimal portfolio with the backtesting method.