Abstract Forecasting the volatility of the key climate predictor variables including rainfall and temperature of has been the subject of extensive research among academics, practitioners and meteorology professionals. This thesis estimates a variety of multivariate GARCH models using a monthly time series of rainfall (in mm) and a monthly time series of temperature (in ) and compares the forecasting performance of these models based on a GARCH model selection mechanism, that is, using the Mincer-zarnowitz regression and the forecast error measures, a model with minimum forecast error measure value and maximum MZ value is selected as an appropriate model. We can choose a single bi-variate GARCH model, and then estimate the parameters of the selected model. The analysis points to the conclusion that time varying conditional correlation (VCC) model with Student’s t distributed innovation terms is the most accurate volatility forecasting model in the context of our empirical setting. We recommend and encourage future researchers studying the modeling and forecasting performance of MGARCH models to pay particular attention to the measurement of the hot issue in the world i.e. climate key predictor variables of volatility. Keywords: key climate predictor variables of co-volatility, conditional correlation, forecasting, multivariate GARCH