Based on the tutorial document, I tried to use the df_mutual_information to select the lag value. But I found that the first minimum score values of time series X and Y are different. What can I do next? Using the lag with first minimum score value of X to predict X from Y? And using the lag with the first minimum value of Y to predict Y from X?
Based on the tutorial document, I tried to use the df_mutual_information to select the lag value. But I found that the first minimum score values of time series X and Y are different. What can I do next? Using the lag with first minimum score value of X to predict X from Y? And using the lag with the first minimum value of Y to predict Y from X?