How to Find Maximum Likelihood Estimators for Sample Data?

2 replies · 2K views · last reply
Scootertaj
Messages
97
Reaction score
0
Printed problem from Amemiya p. 157: table giving P(W=w) for W=1 as mu squared, W=2 as 2mu(1-mu), W=3 as (1-mu)squared, with 0 < mu < 1; sample of size 10 with X=1 three times, X=2 four times, X=3 three times; parts a, b, c.

Homework Equations



[tex]L(x,p) = \prod_{i=1}^npdf[/tex]
[tex]l= \sum_{i=1}^nlog(pdf)[/tex]
Then solve [tex]\frac{dl}{dp}=0[/tex] for p (parameter we are seeking to estimate)

The Attempt at a Solution



I know how to do this when we are given a pdf, but I'm confused how to do this when we have a sample.
 
Scootertaj said:
Problem statement: discrete random variable W with a table giving P(W=1) = μ², P(W=2) = 2μ(1−μ), P(W=3) = (1−μ)², where 0 < μ < 1; a sample of size 10 has X=1 three times, X=2 four times, X=3 three times, followed by parts a) to c).



Homework Equations



[tex]L(x,p) = \prod_{i=1}^npdf[/tex]
[tex]l= \sum_{i=1}^nlog(pdf)[/tex]
Then solve [tex]\frac{dl}{dp}=0[/tex] for p (parameter we are seeking to estimate)

The Attempt at a Solution



I know how to do this when we are given a pdf, but I'm confused how to do this when we have a sample.

Are there no similar examples solved in your textbook or course notes? Could you find nothing at all on-line?

RGV
 
Not so far, though I'll be talking to a graduate TA about it.