Enhanced Risk Management: Integrating Fuzzy Logic and EWMA Model for Market Volatility

In the financial markets, volatility is vital for hedging, assets risk management, and the derivatives pricing. Consequently, it is essential to forecast markets volatility accurately. The standard exponential weighted moving average (EWMA) model tracks market shift in conditional variance of returns by prioritizing the recent observations and decreasing weights exponentially to the past returns. We propose a new fuzzy based FEWMA model that integrates fuzzy logic systems, k-means clustering and EWMA model to forecast market volatility more accurately by assigning weights to returns on the basis of fuzzy logical relationships and rules. The proposed model’s performance is compared to that of the classic GARCH, EWMA, and different modified versions of EWMA models in terms of statistical accuracy measures. This research aims to introduce fuzzy theory in RiskMetrics modeling scheme which adopts nonlinear structure to assign suitable weights to past returns. Furthermore, proposed FEWMA model’s structure combines an effective fuzzy logical control system and pattern learning in k-means clustering environment to forecast volatility with higher accuracy.