Abstract
The exponentially weighted moving average (EWMA) model is a particular modelling scheme advocated by RiskMetrics that is capable of predicting the current level of financial time series volatility. It is designed to track changes in conditional variance of financial returns by assigning exponentially decreasing weights to the observed past squared measurements.
Recently, a recursive estimation technique suitable for this class of stochastic processes has been introduced and discussed. It represents a computationally attractive alternative to the already established non-recursive estimation strategies since it is effective in terms of memory storage, computational complexity and its ability to estimate and control the EWMA modelling scheme in real time.
The aim of the paper is to investigate prediction accuracy of different EWMA model estimators. By analysing a set of eighteen very diverse world stock indices, this study has shown that the recursive estimation scheme can be recommended due to its advantageous properties if predicting the volatility; it is competitive to other approaches commonly used in financial practice.