# How do I write a log likelihood function when I have 2 mean values for my pdf?

I have been given the following pdf :

fT (t; B, C) = ( exp(-t/C) - exp(-t/B) ) / ( C - B ) , (t>0) where the overall mean is B+C.

I am unsure as to how to write the log likelihood function of B and C.

The next part of the Q asks me to derive the equations that would have to be solved in order to find the max likelihood estimators of B and C.

I would be grateful for any help =)

-

By definition, the log-likelihood is given by $$\ln \mathcal{L}(B,C|x_1 , \ldots ,x_n ) = \sum\limits_{i = 1}^n {\ln f(x_i|B,C)}.$$ Thus, in our example, $$\ln \mathcal{L}(B,C|x_1 , \ldots ,x_n ) = \sum\limits_{i = 1}^n {\ln \bigg[\frac{{e^{ - x_i /C} - e^{ - x_i /B} }}{{C - B}}\bigg]} .$$
EDIT: In view of the next part of the question, it may be useful to write $$\ln \mathcal{L}(B,C|x_1 , \ldots ,x_n ) = \sum\limits_{i = 1}^n {\ln [e^{ - x_i /C} - e^{ - x_i /B} ]} - n\ln (C - B).$$
The likelihood function is given by $\mathcal{L}(B,C|x_1 , \ldots ,x_n ) = \prod\nolimits_{i = 1}^n {f(x_i |B,C)}$. – Shai Covo Jan 2 '11 at 22:06
Wouldn't it be enough to write the equations corresponding to $\frac{\partial }{{\partial C}}\ln \mathcal{L}(B,C|x_1 , \ldots ,x_n ) = 0$ and $\frac{\partial }{{\partial B}}\ln \mathcal{L}(B,C|x_1 , \ldots ,x_n ) = 0$? – Shai Covo Jan 2 '11 at 22:44