Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $Y_1, Y_2, \ldots, Y_n$ iid random variables with density $f(y)=\theta\cdot y^{\theta-1}$, $0<y<1$, $\theta >0$.

I need to show that the maximum likelihood estimator of $\theta$ is $-n\bigm/\sum\ln Y_i$.

I know the first step is to get the likelihood of $\theta$

$$L(\theta) = \theta\cdot 1^{\theta-1} \cdot \theta\cdot 2^{\theta-1} \cdots \theta\cdot n^{\theta-1} $$ But I am at a loss of how to simplify this; I keep getting scratches of dead-end work... (It is something wrong with my algebra or partial differentiation).

share|improve this question
    
That is not at all the likelihood! –  Bunder Jul 9 '13 at 6:41

1 Answer 1

Your likelihood is not correct; it should be $L(\theta) = (\theta y^{\theta-1}_1) \cdot (\theta y^{\theta-1}_2) \cdot \ldots \cdot (\theta y^{\theta-1}_n)$, which is $\theta^n (\prod^n_{i=1} y_i)^{\theta-1}$. Then $\ln(L(\theta)) = n \ln \theta + (\theta-1) \sum^n_{i=1} \ln y_i$. Differentiating this gives $n/\theta + \sum y_i$. Equating this to zero gives $\hat{\theta} = -n / \sum \ln y_i$.

share|improve this answer

Your Answer

 
discard

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.