Forecasting the Monthly Real Wage Rate of the Public Sector in Sri Lanka

This study assesses the monthly real wage rate of the public sector of Sri Lanka by elaborating a suitable time series model to identify the future trends associated with the real wage rates of Sri Lanka. The sample data set consists of monthly real wage rate data from January 2018 to March 2024 from the Central Bank of Sri Lanka (CBSL). The real wage rate has been calculated selecting 2016 as the base year. Suitable parsimonious models were identified through the patterns of the sample parti al autocorrelation function (PACF) and sample auto-correlation function (ACF) of the stationary series. Based on the indications such as Akaike information criterion (AIC), Schwarz Criterion (SC) and log likelihood an autoregressive integrated moving average (ARIMA) model of the type (0,1,2) was distinguished as the best fitted model. The residuals of the best fitted model were ascertained to be white noise. The model has been validated for the first three months of 2024. The Mean Absolute Percentage Error (MAPE) for the validation data is 9.59%. The forecasted wage rate values from April 2024 to September 2024 are 54.562, 54.096, 53.631, 53.165, 52.7 and 52.234 respectively. The study’s findings can be utilized by policymakers, economists, and government workers to improve their financial planning. Keywords: ARIMA models; Box Jenkins Methodology; Forecasting; Real wage rate; Time series analysis