VaR and time‐varying volatility: a comparative study of three international portfolios

Abstract

Purpose -This study aims to evaluate the market risk exposure of three international equity portfolios using value-at-risk (VaR). This risk metric calculates the worst case loss for a business in the course of its daily transactions. To ensure that the calculated VaR reflects emerging risk characteristics, this paper introduces an approach that incorporates time-varying volatility. Design/methodology/approach -This study uses the GARCH technique to calculate the volatility metric with which VaR estimates are obtained. The out-of-sample performance of the VaRs is then assessed by comparing them to the actual market risk losses in that period. Findings -Empirical results show that regardless of market conditions, the VaR calculated with this (GARCH) approach is more robust and more reliable than the traditional methods. Pursuant to the banking regulation on market risk capital stipulated by the Basel Committee on Banking Supervision, the out-of-sample VaRs are at least equal to actual daily market risk losses at the 99 percent confidence level. Practical implications -The key goal of banking regulation is to ensure that financial firms have sufficient capital for the types of risks they take. Determining the right amount of capital requires these firms to first estimate their worst case loss, which is the value-at-risk. The approach to the calculation of VaR introduced in this paper enhances the accuracy in the measurement of market risk capital for financial institutions. Originality/value -This paper recognizes that for VaR to fully account for market risk losses, the risk metric must be correctly measured. The unparalleled approach in this paper of incorporating time-varying volatility in VaR calculations offers banking institutions a more reliable means of determining their capital adequacy.

Value at Risk in the Formation of Optimal Portfolio on Sharia-Based Stocks

“…(7) = Standard deviation of the portfolio = Stock return variance i = Covariance between stocks i and j = Proportion of funds invested in stocks i = Proportion of funds invested in stocks j Double addition marks, meaning will be added together n = number of stocks in the portfolio 8 [20] 1. Determine the parameter values for portfolio returns and correlations between variables (stock returns).…”

Retail Price Volatility: The Case of Coconut in Sri Lanka

“…This model is broadly applied and therefore, valuable to review thoroughly. According to Obi and Sil (2013), the basic GARCH model for the superb forecaster of the next period's variance (𝜎 𝑡+1 2 ) is that the weighted average of the long term variance, the constant term (ω), new information about volatility observed in the current period, the ARCH term (𝜀 𝑡 2 ), and the forecast variance for the current period, the GARCH term (σ2 t ). Hence, the standard specification of GARCH (p, q) is symbolized as follows.…”