I am slightly getting confused at the presentation of regression models.

What would be the difference between these two:

$y = \beta_0 + \beta_1 x_1 + \beta_2 x_2$

$y = \gamma_0 + \gamma_1 x_1 + \gamma_2 x_2$

When are the different notations to be used?

I'm unable to search this concept well, so it's quite confusing as the literature seems to use the symbols on similar occasions.

Is it just a different symbol for the same thing or is there a specific explanation as to when each notation should be used?

  • 5
    $\begingroup$ Different authors might use different notations, but the choice of symbols is just a matter of taste (or a convention by field). $\endgroup$ – QuantIbex Aug 5 '13 at 13:32
  • $\begingroup$ @Penguin_Knight seems to suggest otherwise in this post $\endgroup$ – Cesare Camestre Aug 5 '13 at 13:40
  • $\begingroup$ What makes you think that? He might have used $\gamma$s instead of $\beta$s in his first model, and $\beta$s instead of $\gamma$s in his second model. $\endgroup$ – QuantIbex Aug 5 '13 at 13:47
  • 2
    $\begingroup$ Although there are some symbolic conventions in statistics (principally for names of probability distributions), by and large statistical notation uses mathematical rules: in that light, your question is the same as asking whether it makes any difference whether you name the unknown in an algebra problem "x" or "y". Obviously that is immaterial. What is of import is how the names function in any expression. That function should be adequately described in the surrounding narrative--if not, you should leave a comment asking for clarification. $\endgroup$ – whuber Aug 5 '13 at 16:36
  • 1
    $\begingroup$ In particular contexts, γs instead of βs might carry a particular meaning, but in general (since you didn't supply a particular context), it's simply a matter of either convenience, or convention. $\endgroup$ – Glen_b Aug 6 '13 at 1:20

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.