0
$\begingroup$

I have read a lot of questions and answers but I am still confused for my case. My dependent variable (employment growth) is measured : $$ \frac{((number\ of\ employees \ in \ year\ t) - (number \ of \ employees \ in \ year \ t-1))}{[((number \ of \ employees \ in \ year \ t) + (number \ of \ employees \ in \ year \ t-1))/2]}*100 $$ It is defined in percentage (%). It can takes values from -100 to 100 (continuous). 1)I have an independent in log value. 2) I have an independent - indicator which can take values from 0 to 50 (continuous). Let's say I have this: $$ eg = 80 + 1.48 \ log(x) + 6.044 \ Indicator + \epsilon $$ How I interpret coefficient for each independent... Thanks in advance!

$\endgroup$
2
  • $\begingroup$ What about it confuses you? Have you taken a course in regression? $\endgroup$
    – Peter Flom
    Mar 18 '18 at 13:38
  • $\begingroup$ So, my point is : 1) For a 1% increase in x , we expect a 1.48 % increase in eg. 2) For a 1-unit increase in indicator, we expect a 6.044 increase in eg. Am I right ? The thing is that eg is already defined in percentage (%) , so I think there is something wrong with my thoughts... $\endgroup$
    – ec on
    Mar 18 '18 at 13:47
0
$\begingroup$

You can interpret it like a regular regression (as you did in your reply to my comment, above). However, since the dependent variable is bounded by -100 and 100, regular OLS regression might not be appropriate and the regression equation might make nonsensical predictions. This thread or this one may be helpful.

$\endgroup$

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.