0
$\begingroup$

My time series data looks like random generation which was obtained by aggregation day-wise sales of a product. The dependent variable does not show any kind of pattern.

So have tried tries different methods like data standardization, Lagged features, Rolling window statistics through xgb boost regressor,but have not got good results.

Could you please suggest what other feature techniques can be applied to my TS data.

Sample Data

 Sale_date  Sales
2010-03-31  4.0
2010-04-02  2028.0
2010-04-03  1865.0
2010-04-04  1080.0
2010-04-05  1560.0
2010-04-06  214.0
2010-04-07  33.0
2010-04-09  300.0
2010-04-10  22.0
2010-04-12  250.0
2010-04-13  88.0
2010-04-14  10.0
2010-04-16  410.0
2010-04-18  57.0
2010-04-20  70.0
2010-04-21  3020.0
2010-04-24  44.0
2010-04-25  83.0
2010-04-26  664.0
2010-04-27  2580.0
2010-04-28  501.0
2010-05-02  100.0
2010-05-03  50.0
2010-05-04  286.0
2010-05-05  20.0
2010-05-08  144.0
2010-05-11  341.0
2010-05-12  16.0
2010-05-15  92.0
2010-05-16  1320.0
2010-05-19  89.0
2010-05-21  126.0
2010-05-22  465.0
2010-05-24  72.0
2010-05-25  1475.0
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.