
Alan C. answered 05/11/20
Math Tutor in Central NJ
To find the maximum likelihood estimator, we need to maximize the likelihood function, which usually consists of three things:
1) find the likelihood function
2) find the derivative of the likelihood function with respect to any parameters (in this case, θ)
3) set the derivative equal to zero
With the occasional intermediary step of taking the natural log of the likelihood function after finding it, since the natural log is a monotonic transformation, and the log of the likelihood function will have the same maximum as the original likelihood function.
Assuming
f(x,θ)=(x3) . (e-x/θ) / 6θ4
Our likelihood function is f(x1,θ)*f(x2,θ)*f(x3,θ) since we have three observations
We next take the log of this to get:
ln[(x13) . (e-x1/θ) / 6θ4]+ln[(x23) . (e-x2/θ) / 6θ4]+ln[(x33) . (e-x3/θ) / 6θ4]
=3ln(x1)+-x1/θ-ln(6)-4ln(θ)+3ln(x2)+-x2/θ-ln(6)-4ln(θ)+3ln(x3)+-x3/θ-ln(6)-4ln(θ)
We take the derivative of this with respect to θ, and get
df/dθ=x1*θ-2-4/θ+x2*θ-2-4/θ+x3*θ-2-4/θ
And set this equal to zero and solve for θ
0=x1*θ-2-4/θ+x2*θ-2-4/θ+x3*θ-2-4/θ
Our final answer is
θ=1/12*(x1+x2+x3)
Hope this helps!