0
$\begingroup$

Data

  • Amount spent on product type A
  • Amount spent on product type B
  • 5 age groups
  • External factor X
  • External factor y

I have data in the form of monthly sums over 60 months for all of those data-points summarized above.

Goal

  • Understand how external factors X and Y affect the propensity of people in 5 age groups to spend on product A.
  • Understand how external factors X and Y affect the propensity of people in 5 age groups to spend on product A.

Questions

  • Is multivariate regression an efficient and effective means of identifying relationships?
  • Can I run this test in Excel? If not, is there a simple means of doing this on R or some other free software?
Improve this question
$\endgroup$
1
  • $\begingroup$ The term multivariate regression is best used to mean a model with two or more response variables. As you are interested in predicting sales of Product A, that could be a problem using multiple regression. Increasingly, the prefix multiple is fading away, even when there are several predictors. It used to be a big deal to do multiple regression, but that was 50 years or more ago. $\endgroup$
    – Nick Cox
    Commented Apr 14, 2020 at 13:25

1 Answer 1

Reset to default
1
$\begingroup$

Given the little detail about what the data represents and how it was collected the most probable answer is: Yes, you can try to predict the influence each age group and each factor has on either A or B.

While running linear regression in Excel is certainly possible I advise to have a look into JASP software. JASP is free software. It will integrate well with Excel. Linear regression and the associated tests will be done in R under the hood but your experience will be a clean graphical user interface combined with print-ready output.

Start at https://jasp-stats.org/getting-started/

Improve this answer
$\endgroup$
2
  • $\begingroup$ Thank you for your input. I have historical data on stock volumes and futures volumes traded by clients in various age groups. I'm looking to understand how the currency exchange rate and how the general stock market index affect or mirror the amounts invested by people in each age group. I may have been needlessly oblique. $\endgroup$
    – NuStack
    Commented Apr 14, 2020 at 22:54
  • $\begingroup$ Please do not get me wrong -- all that sounds as if linear regression could be the most obvious starting point but the decision on whether linear regression is suitable depends on detail. Go for linear regression and stay away form computing statistics in Excel - even if it may be the best software for other jobs. If analysing data is your job, the investment in learning R is probably worth it but just to get your toes wet, JASP, PSPP, R Commander and others are good. $\endgroup$
    – Bernhard
    Commented Apr 15, 2020 at 6:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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