The study is designed to forecast, production of rapeseed and mustard in Pakistan by using time series analysis. Among different appropriate models like (1,0,0), (1,0,1), and (2,0,0) it is based on Normalized Bayesian criterion (NBIC), Akaike information criterion (AIC), Hannan-Quinn and Schwarz Criterion that Autoregressive (AR) model (1,0,0) is best. The Brock, Dechert and Scheinkman (BDS) test is also applied on the residuals of model (1,0,0) for checking the linearity condition before forecasting. The forecasted values for the year 2016-17 to 2021-22 indicate an increasing trend in future of rapeseed and mustard production in Pakistan.