I have a time series data of monthly sales for the last 4 years. The sales for one month eg Feb 2017 is much higher same month last year or the last month sales of Jan 2017.

If i wanted to detect if this was only due to random chance which test is advised



Identify an ARIMA model using all the data and then add an indicator series for the last point (0,1) to your model. Compute the t value associated with that pulse indicator. Convert the t value to a p value and you have your answer. If the value you are trying to test is the last value then you might exclude it from the ARIMA identification process.

  • $\begingroup$ Could you elaborate please...on the steps required $\endgroup$ – Mazen Aug 19 '17 at 12:53
  • $\begingroup$ I am not sure what more i can write .Perhaps you can either call me or contact me via SKYPE and I will try and verbalize the steps. $\endgroup$ – IrishStat Aug 19 '17 at 13:33

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.