While working on open source Time Series Data for Grocery sales where my aim is to perform Demand forecast of Grocery Products, I completed the cleaning of data of Grocery sales data, then in order to check whether the Grocery dataset is Stationary or not, I conducted Dickey Fuller statistical test on Historical Sales column of Grocery Dataset
and got result that data is Stationary, Pls refer below output of Python code :-

ADF Statistic: -29.450181 p-value: 0.000000 Critical Values: 1%: -3.710 5%: -2.971 10%: -2.678 Reject Ho - Time Series is Stationary

Now when i visualize the historical Sales on which i conducted Statistical Dickey Fuller Test which gave result as Time Series is Stationary,But Visually it is giving output like a Curve going up and down at regular intervals demonstrating Seasonality.

Visually it shows seasonality but when conducted statistical test it says time series data is stationary, so can i proceed ahead without removing Seasonality as i will be using Regression based Models and not Time Series Models to predict the Demand Forecast of Grocery Products.

Please Advice.

  • $\begingroup$ Hi. Can you please put your main question in the title? $\endgroup$
    – nbro
    Aug 17 '20 at 23:51
  • $\begingroup$ Hi nbro, Do you want me to Modify the existing Title of the question which i have asked ? Please Advice $\endgroup$
    – amrev15
    Aug 19 '20 at 18:20
  • $\begingroup$ Updated the Main Question, Please advice regarding my query. $\endgroup$
    – amrev15
    Aug 23 '20 at 7:40
  • $\begingroup$ It is better to trust your eyes than trusting some tests you have understood poorly! You probably needs a seasonal model. Augment your post with plots of the series! $\endgroup$ Feb 3 at 1:24

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.