0
$\begingroup$

I am working on a task to predict the sales of all the items offered in all the countries. The sales are aggregated on a daily and country level. Each Item has a history of past sales and prices for a given country. I want to predict the sales in upcoming n days.

Items have different sized histories, with some items being on sale for months, whereas others have been on sales for just a couple of weeks. Also note that during data analysis i noticed that price influenced the sales, when it is reduced customers bought more items of that type.

I just want to know the approach in predicting the sales. How should i approach in predicting the item wise sales and the effect of reducing the price. What should be my path to follow ?

$\endgroup$
0
$\begingroup$

If you have daily data that includes period of time where no sales have occurred , the approach is called Intermittent Demand . If you have daily data (for each country ) that is generally non-zero then you should be looking at models that include anthropomorfic factors such as day-of-the-week, lead and lag effects of holiday indicators , week-of-the-month , day-of-the-month , monthly effects while incorporating user-specified predictor series like price or promotion effects AND any memory effects that might be present ALONG with possible trends and or level/step shifts.

An example of this can be found at An example of this kind can be found at https://autobox.com/capable.pdf which I have helped to develop starting at slide 49 which may be of some help to you. Additionally search for daily data in SE and review some discussions.

$\endgroup$

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.