Optimal asset allocation under var criterion: Taiwan stock market

Abstract

Value at risk (VaR) measures the worst expected loss over a given time horizon under normal market conditions at a specific level of confidence. These days, VaR is the benchmark for measuring, monitoring, and controlling downside financial risk. VaR is determined by the left tail of the cumulative probability distribution of expected returns. Expected probability distribution can be generated assuming normal distribution, historical simulation, or Monte Carlo simulation. Further, a VaR-efficient frontier is constructed, and an asset allocation model subject to a target VaR constraint is examined.This paper examines the riskiness of the Taiwan stock market by determining the VaR from the expectedreturn distribution generated by historical simulation.Our result indicates the cumulative probability distribution has a fatter left tail, compared with the left tail of a normaldistribution. This implies a riskier market.