This study highlights application of different Autoregression models for time-series forecasting of the Net Asset Value (NAV) of Axis Retirement Savings Fund stock. Autoregression models are widely used for time-series prediction. The primary aim of this study is to develop a machine learning model for precise prediction of stock market NAV thereby mitigating the associated financial risks of the investor. The machine learning models considered in this study were persistence models and fixed autoregression models. The historical data of the previous 100 days were considered for training of the proposed models using the Scientific Python Development Environment (SPYDER) platform. The performance of the model was tested for 7 days ahead predictions. The performance of the models was compared in terms of mean absolute error (MAE), root mean squared error (RMSE) and mean absolute percentage error (MAPE). The findings clearly highlight that the fixed autoregression model outperforms the persistence model giving a lower MAPE responses in order of 0.445l4%.