search

DescartesResearch / telescope

Telescope is a hybrid multi-step-ahead forecasting approach based on time series decomposition.
GNU General Public License v3.0
11 stars 6 forks source link

Train recommender with the same data fitted to the prediction #20

Open nateehuang opened 4 years ago

nateehuang commented 4 years ago

Thank you for sharing your insights and codes. I am trying to apply this framework in financial time series and wonder if it is valid to use the same data used in training the forecast model to train the recommender.

DrAndreBauer commented 4 years ago

Hello @nateehuang

it depends on what you want. If you use the same time series data set for the training of the recommender and for the forecast, the recommendation will have a small selection error, so that for the forecast the best method of machine learning is most likely used. In other words, if you want good forecast results, then it is valid :)