Really nice work. I have question concerning the data pre-processing. The accuracy of the model trained on the data which has gone through a 10 day moving average has improved dramatically. However, is data contamination coming from the moving average affecting the prediction? In reality we would not be able to perform a moving average on future stock prices. I wonder if doing a rolling average separately on train, test and validation gives still the same level of accuracy. Thanks!
Really nice work. I have question concerning the data pre-processing. The accuracy of the model trained on the data which has gone through a 10 day moving average has improved dramatically. However, is data contamination coming from the moving average affecting the prediction? In reality we would not be able to perform a moving average on future stock prices. I wonder if doing a rolling average separately on train, test and validation gives still the same level of accuracy. Thanks!