Consider the following method of estimating λ for a Poisson distribution. Observe that
p0 = P(X = 0) = e(-λ)
Letting Y denote the number of zeros from an i.i.d. sample of size n, λ might be estimated by
λ˜ = − log(Y/n)
Use the method of propagation of error to obtain approximate expressions for the variance and the bias of this estimate. Compare the variance of this estimate to the variance of the mle, computing relative efficiencies for various values of λ. Note that Y ∼ bin(n, p0).
Let Y denote the number of zeros from an i.i.d. sample of size n, then
Therefore
Let
The approximate expression for the variance of f(X1,X2,...,Xn) is
since Xi's are independent.
Therefore
,
where
where y is the observed value of Y.
Bias=
Therefore,
.
= holds if n=y i.e. all n samples take the value zero.
Consider the following method of estimating λ for a Poisson distribution. Observe that p0 = P(X...
The number of medical emergency calls per hour has a Poisson distribution with parameter λ. Calls received at different hours are considered to be independent. Emergency calls X1 ,…, Xn for n consecutive hours has the same parameter λ. a) What is the distribution of Sn = ∑ Xi ? b) Provide Normal approximation for the distribution of Sn . c) Provide maximum likelihood estimation of λ. Calculate variance and bias of MLE. d) Calculate Fisher information and efficiency of...
ANSWER QUESTION 2
1. The size of claims made on an insurance policy are modelled through the following distribu- tion: λ+1 You are interested in estimating the parameter λ > 0, using the following observations 120, 20, 60, 70, 110, 150, 220, 160, 100, 100 (a) Verify that f is a density (b) Find the expectation of the generic random variable X, as a function of when > 1 (c) Prove that the method of moments estimator of λ is...
Return to the original model. We now introduce a Poisson intensity parameter X for every time point and denote the parameter () that gives the canonical exponential family representation as above by θ, . We choose to employ a linear model connecting the time points t with the canonical parameter of the Poisson distribution above, i.e., n other words, we choose a generalized linear model with Poisson distribution and its canonical link function. That also means that conditioned on t,...
As on the previous page, let Xi,...,Xn be i.i.d. with pdf where >0 2 points possible (graded, results hidden) Assume we do not actually get to observe X, . . . , Xn. to estimate based on this new data. Instead let Yİ , . . . , Y, be our observations where Yi-l (X·S 0.5) . our goals What distribution does Yi follow? First, choose the type of the distribution: Bernoulli Poisson Norma Exponential Second, enter the parameter of...