1

I was wondering if there is any limitation regarding the number of independent variables if I want to simultaneously include all variables in one regression model.

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
Hooman
  • 11
  • 3

1 Answers1

0

Not in principle, but there will be practical limits based on computing power, sample size and multiple testing issues.

Arne
  • 203
  • 2
  • 4
  • 1
    What happens when the number of independent variables exceeds the number of observations? – whuber Mar 13 '20 at 13:07
  • 1
    Then you have a choice between infinitely many model instances that all fit the data perfectly. I suppose you could call that a principled limitation. – Arne Mar 13 '20 at 13:41
  • 1
    Be a little cautious: having more variables than observations does not imply the regression will fit perfectly! – whuber Mar 13 '20 at 13:43
  • Yes, not in general, I was thinking of linear regression only. – Arne Mar 13 '20 at 13:53
  • This has been discussed here: [modelling with more variables than data points](https://stats.stackexchange.com/questions/223486/modelling-with-more-variables-than-data-points) – Arne Mar 13 '20 at 13:56
  • How can we calculate number of observations in linear regression using SPSS? – Hooman Mar 13 '20 at 14:31
  • If your data are properly formatted, the number of observations should be equal to the number of rows in your SPSS table, so you can just look at the number of the last line. – Arne Mar 13 '20 at 15:00
  • Could you please talk more specifically? which line and rows are you referring to? – Hooman Mar 13 '20 at 15:20
  • 3
    *Especially* in linear regression, having more variables than observations is no guarantee of a perfect fit. As an example, suppose all the variables are multiples of each other. The point is that the proper measure of how many variables there are is not their count: it's the dimension of the vector space they span. – whuber Mar 13 '20 at 15:38