Likelihood function for data y under given distribution with probability p

31 Views Asked by At

I am struggling with the following question:
Consider a random variable (RV) Y that follows the distribution: $$P(Y=y|p) = (1-p)^yp$$ where $y \in$ {0,1,2,..} is a non-negative integer, and p is the parameter of the distribution. Imagine we observe a sample of n non-negative integers y={$y_{1},....y_n$) and want to model them using the given distribution ( the data is independently and identically distributed).

Write down the likelihood function for the data y ( i.e, the joint probability of the data under the given distribution with the probability parameter p )

I have got as far as the following ( I think) but am lost on the next step.

$\prod_{i=0}^y p \prod_{i=0}^y(1-p)^y$

I have tagged maximum-likelihood because there is no likelihood tag.

1

There are 1 best solutions below

7
On

The likelihood function of the parameter $p$ given the data $(y_1,\cdots,y_n)$ is $$ L(p\mid y)=P(Y_1=y_1,\cdots, Y_n=y_n\mid p)=\prod_{i=1}^n P(Y_i=y_i\mid p) $$ where $Y_i$'s are iid copies of $Y$.

Now you simply do the substitution.