I am trying to find a macroeconomic model that fits my data mainly using simple multiple regression.

However, I am a little confused with the mistakes I may be doing. The question below might sound dumb for most of you guys, but for me it's tricky.

The mainly confusion is if I have to use the % variation to compare the data or if I can use the Gross Data.

For example, let's say I am trying to find a model to predict sales based on my historical data. I have the number of t-shirts I sold each quarter, I have the GDP of my country, the Price Index (inflation) and Consumer Confidence. Is it a problem if I run a regression using the number of t-shirts I sold instead of the growth/decrease of them? And the same question I don't know how to answer when talking about the GDP.

My first instinct is to use the variation. Since we can have some sort of scale problem, but again, I'm not so sure about that. What do you guys think? I will be making any mistake if I consider the gross data for obtaining the correlation and the multiple regression model?


It appears you have time series data and standard multiple regression is not the preferred approach due to time series complications.This answer Appropriate predictive model for two random time series with serial correlation might be of interest to you. The difference between regression an multivariate Box-Jenkins is discussed here in a blog that I wrote http://www.autobox.com/cms/index.php/afs-university/intro-to-forecasting/doc_download/24-regression-vs-box-jenkins .

| cite | improve this answer | |
  • $\begingroup$ That might work.. I will take a look at both links you sent me, many thanks! $\endgroup$ – trder Aug 6 '15 at 23:46

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.