Accuracy in risk estimation based on simple SMA and EWNA models: Evidence from Macedonain Stock Market

Risk estimation or volatility estimation at financial markets, particularly stock exchange markets, is complex issue of great importance to theorists and practitioners. Models used to estimate volatility forecasts are translated into better pricing of stocks and better risk management. The aim of this research is to test applicability of simple models like Simple Moving Average (SMA) and Exponentially Weighted Moving Average (EWMA) to estimate risk. The performance of SMA and EWMA with rolling window of 100 using 0.94, 0.96, and 0.90 as smoothing constant were analyzed on investment activities of time series of 10 stocks comprising MBI-10. Binary Loss Function (BLF) is employed to measure accuracy of VaR calculations, because VaR models are useful only if they predict future risks accurately. Results show that risk managers can use SMA (100) and risk metric EWMA(100) smoothing constant of 0.96 model as a tool for estimating market risk at 95% confidence. At 99% confidence level both models failed to estimate risk accurately and permanently underestimate the risk.