Volatility forecasting approach to risk assessment of private equity mutual funds in Malaysia

Volatility, though unobservable and latent in nature, is forecastable due to its persistency over time. Financial volatility measures the risk of financial assets’ returns. Voluminous literatures on volatility forecasting studies imply risk assessment through volatility study is a pre-requisite in managing risk of financial assets. Stocks volatility studies have well contributed to the financial volatility literatures but not mutual funds volatility, typically in the case of Malaysia. Inconsistency between funds’ objectives and their risk-return relationship attributed to fund managers inability to time the market and stocks selectivity over different business cycles as documented in past studies, substantiate the risk involved in mutual funds investment. The key motivation of the study is to examine risk of mutual funds investment through volatility forecasting approach. The approach of this study focussed on developing and determining appropriate forecasting models by way of forecasting accuracy comparison. Daily return of seven fund indices (Growth, Growth & Income, Income, Balanced Growth, Balanced Growth & Income, Balanced Income and Mixed Asset Growth) generated from 57 private equity mutual funds of different investment objective and corresponding risk-return characteristics across two sub-periods “with financial crisis” (2005-2011) and “without financial crisis” (2012-2019) are examined. The empirical evidence from GARCH in Mean (GARCH-M) revealed existence of inconsistency between fund objectives and risk-return relationship across seven fund indices. Funds’ return volatility is found to be more volatility in sub-period with financial crisis. The asymmetric EGARCH under Student-t distribution captures the asymmetrical leverage effect well and emerge as the best GARCH model. However, the robust and outlier resilience STES with Error and Absolute Error transition variable is the overall best model in the one-day ahead volatility forecasting. The Realized Variance MIDAS (RVar-MIDAS) model outperformed both the STES methods and the GARCH models in longer lead time (one-week ahead) forecasting. Results from the GARCH-MIDAS revealed macroeconomic variables (output, inflation, interest rate, money supply and exchange rate) examined exert small impact on funds volatility while realized volatility exerts stronger impact, implying sensitivity of mutual funds volatility to financial or macroeconomic news than changes of the macroeconomic variables. Findings of this study has contributed empirical forecasting models in assessing risk of mutual funds investment using the mutual funds indices generated allows a macro-analysis of funds risk. Inconsistency between funds objectives and their risk-return relationship warrants tighter monitoring of fund managers trading behaviour by Securities Commission of Malaysia. The sensitivity of funds volatility to both financial and macroeconomic news provides a direction to the Malaysian government to exercise cautious execution of macroeconomic policies so as not to trigger unnecessary volatility in the Malaysian financial market.

Text SPE 2021 20 IR.pdf Download (2MB)

Actions (login required)

View Item