0

I was wondering if it is possible to state that FDI-Foreign Direct Investments, per capita is a count variable?

By definition count variable is number of accidents, number of suicides.

So theoretically FDI per capita is also considered to be a count variable, since it is a number of FDI per capita.

Correct me if I am wrong.

kjetil b halvorsen
  • 77,844
Romas
  • 29
  • 1
    What is FDI? Please explain ... – kjetil b halvorsen Apr 19 '18 at 10:15
  • No; clearly you can state that, but I think it would generally be considered quite wrong. once you divide then non-integer values are not only possible but usual. The key is are the data you are using discrete and integers $\ge 0$ as a matter of principle? No model knows where they came from. An exception could be any model where numerator and denominator are specified separately. – Nick Cox Apr 19 '18 at 10:18
  • FDI - Foreign Direct Investments.

    Yes, the data I am using is discrete and all integers are above or equal to zero.

    The reason I was wondering if FDI per capita is a count variable, I want to use it as a dependent variable in negative binomial regression. Do you think it would correct to do so?

     – Romas Apr 19 '18 at 10:45
  • Couldn’t you end up with the numerator not being an integer multiple of the denominator e.g. $\frac{17}{4}?$ – Dave Oct 03 '20 at 18:39

1 Answers1

1

Maybe you are counting the dollars invested, but I would not take that as a count variable! Or maybe you are counting discrete investment projects irrespective of size, but I doubt that is the usual meaning of FDI. So no, FDI is not a count variable.

Nick Cox
  • 56,404
  • 8
  • 127
  • 185
kjetil b halvorsen
  • 77,844