MODELLING TAIL BEHAVIOR OF RETURNS USING THE GENERALIZED EXTREME VALUE DISTRIBUTION
MONNYE RHODA MAKHWITING, CASTON SIGAUKE, MASEKA LESAOANAABSTRACT. Modelling tail behavior of rare and extreme events is an important issue in the risk management of a financial portfolio. Extreme Value Theory (EVT) provides the essentials needed for the statistical modeling of such events and the computation of extreme risk measures. The modeling of extreme daily share returns at the Johannesburg Stock Exchange (JSE) over the period 2002 to 2011 is discussed in this paper. Stock returns at the JSE market are highly volatile and non-normal. Parameters of the Generalized Extreme Value (GEV) distribution are estimated by Maximum Likelihood Estimation (MLE). Empirical results show that the Weibull distribution can be used to model stock returns on the JSE. pp. 41–52
JEL Codes: D81; G32
Keywords: GEV distribution; MLE; Tail quantiles; Return level
How to cite: Makhwiting, Monnye Rhoda, Caston Sigauke, and Maseka Lesaoana (2014), “Modelling Tail Behavior of Returns Using the Generalized Extreme Value Distribution,” Economics, Management, and Financial Markets 9(1): 41–52.