Time series analysis of settlement prices for individual currency futures in Singapore

Abstract

This paper investigates the efficiency of the currency futures market in the Singapore International Monetary Exchange. The weak sense of market efficiency is tested, with the random walk model being used as the benchmark for comparing univariate models fitted to the three major currency futures, namely deutschmark. Japanese yen and British pound. In weak-form tests of the efficient market hypothesis (EMH), security prices reflect fully all available information based on past values of price data. This means that the weak form tests whether all information contained in the historical prices is fully reflected in current prices. A restrictive version of the weak form of the EMH is the random walk model, which assumes that successive returns are independent and identically distributed over time. Thus, evidence supporting the random walk model is evidence supporting the weak form efficiency of the EMH. Univariate modelling of the data involves fitting several moving average, autoregressive and autoregressive moving average specifications. Using the mean absolute error (MAE), the performances of the estimated models are compared against the random walk model. The three currency futures models consistently outperform the random walk model on the strength of the MAE, which challenges the EMH in the currency futures market in Singapore.