FORECASTING INTERBANK OFFERED RATE USING TIME SERIES APPROACH: SRI LANKAN CONTEXT

Forecasting short term interest rates is a great concern in financial market. Interbank offered rate is a short term interest rate which is used for transactions between commercial banks [1]. In Sri Lanka no prior research has been done to forecast the Sri Lankan Interbank Offered Rate. This study is an attempt to build a forecasting model for Sri Lankan Interbank Offered Rate using time series approach. In literature the interbank offered rates are forecasted using both linear and non-linear time series models. In this study daily overnight Sri Lankan Interbank Offered rate for five years are used. Since it is identified that the time series has non constant variance, different non-linear time series models such as ARCH, GARCH, IGARCH and EGARCH models are fitted and compared them using Akaike Information Criterion and Schwarz Information Criterion values. Past lagged observations of the data series and moving averages are employed as the explanatory variables. It is found that GARCH(1,1) model is the most appropriate with lower accuracy measures to forecast the Sri Lankan Interbank Offered Rate.