log likelihood positive

[GUIRA Asmo]

I used a stochastic frontier model and the log likelihood is positive. I would like to know what explains this result.

[Joao Santos Silva]

Dear GUIRA Asmo,

I do not think that there is much to explain: densities are not bounded between 0 and 1, so the log-likelihood is often positive when the dependent variable is continuous.

Best wishes,

Joao

[GUIRA Asmo]

hello Silva,
thank you very much for your help. by the way, my thesis director has asked me to explain why this is so.

[Nick Cox]

In addition, multiplicative constants are often omitted in calculating and reporting log-likelihood as being immaterial to the position of the maximum.

[GUIRA Asmo]

ok Nick,
This explains the negavity of the log. ok thanks

[Nick Cox]

Sorry, but what is puzzling? Positivity or negativity?

[GUIRA Asmo]

Exuse me it is the fact that the log lik is positive

[Nick Cox]

OK, fine.