Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Trying to find the maximum of a log-likelihood, for a parameter in a covariance function.

I end up with the following problem, that should be concave if my calculations are correct,

\begin{align} &\max_{\kappa}\; -\sum_{i=1}^n \log \left( \lambda_i + \kappa \right) - \sum_{i=1}^n \frac{c_i}{(\lambda_i + \kappa)^2}, \\ &\mbox{subject to }\kappa>0, \end{align}

where $\lambda_i>0,c_i>0$ for all i.

Does there exists a closed form solution?, if not can one find an some bound on the solution?

share|cite|improve this question

The derivative of this function is positive for all $\kappa>0$, so the maximum is taken for whatever maximum you have for your range. Given that you have no upper bound for $\kappa$, this is $+\infty$ (although that isn't a "maximum" so much as a supremum.)

share|cite|improve this answer
ooh, your right, I must have made some error in my calculations. Thanks. – Jonas Wallin Dec 28 '12 at 8:23

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.