1
$\begingroup$

I have data for a retail shop for all of last year (well for many years but this is what I will be using) and all of this year current. The entries are the traffic (# of visitors), the sales, the conversion (those who bought as a percentage of the total that visited), the UPT (average units per transaction), and VPT (average value per transaction in dollars). I have this data for daily, weekly and monthly intervals.

I would like to project revenue in to the future while adjusting for decreased traffic as traffic this year has been down.

How do I go about creating a model for this? What software will suffice for this task (is R enough?) and how can I account for that decreased traffic (which I speculate will be the case for the rest of the year).

I imagine this to be a big project so I just need advice on some of the steps I need to take, the direction really. Any resources are appreciated.

Thank you kindly for any help.

$\endgroup$
  • $\begingroup$ The general approach is to form a SARMAX model otherwise known as a Transfer Function which can embody the impact of causals that you suggest , unspecified holiday effects can be detected , while incorporating seasonal factors as well as anomalies. Take a look at this autobox.com/pdfs/SARMAX.pdf and a very recent post ( like moments ago !) analyzing orders stats.stackexchange.com/questions/407800/… . PS the orders analysis didn't include user specified predictors BUT the general approach does. $\endgroup$ – IrishStat May 11 at 20:38
  • $\begingroup$ stats.stackexchange.com/questions/177262/… comments on our building 600,000 (#of stores) by 50 ( #of products) daily retail models using price and weather as predictor/supporting series. $\endgroup$ – IrishStat May 13 at 18:43
  • $\begingroup$ If you post your data in a csv format , I will try and help you further, $\endgroup$ – IrishStat May 18 at 14:30

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.