Suppose n independent, identically distributed observations are drawn from an exponential () distribution, with pdf given by f(x,)=, 0 < x < .
The data are x1, x2, .. , xn
Construct a likelihood ratio hypothesis test of Ho : vs H1: (where and are known constants, with ), where the critical value is taken to be a constant c
Suppose n independent, identically distributed observations are drawn from an exponential () distribution, with pdf given...
The random variables X1, X2, - .. are independent and identically distributed with common pdf 0 х > fx (x;0) (2) ; х<0. This distribution has many applications in engineering, and is known as the Rayleigh distribution. 2 (a) Show that if X has pdf given by (2), then Y = X2/0 is x2, i.e. T (1, 2) i.e. exponential with mean 2, with pdf fr (y;0) - ; y0; (b) Show that the maximum likelihood estimator of 0 is...
Observations X1,..., Xn are independent identically distributed, following the PDF fx:(xi) = 0x8-1, and that 0<Xi <1 for all i. The parameter is an unknown positive number. Find the ML estimator of e
Let X1, X2,.......Xn be a random sample of size n from a continuous distribution symmetric about . For testing H0: = 10 vs H1: < 10, consider the statistic T- = Ri+ (1-i), where i =1 if Xi>10 , 0 otherwise; and Ri+ is the rank of (Xi - 10) among |X1 -10|, |X2-10|......|Xn -10|. 1. Find the null mean and variance of T- . 2. Find the exact null distribution of T- for n=5. We were unable to transcribe this imageWe were...
Let X1, X2, ..., Xn be a random sample from X which has pdf depending on a parameter and (i) (ii) where < x < . In both these two cases a) write down the log-likelihood function and find a 1-dimensional sufficient statistic for b) find the score function and the maximum likelihood estimator of c) find the observed information and evaluate the Fisher information at = 1. f(20) We were unable to transcribe this image((z(0 – 2) - )dxəz(47)...
Let X1,X2,...,Xn denote independent and identically distributed random variables with variance 2. Which of the following is sucient to conclude that the estimator T = f(X1,...,Xn) of a parameter ✓ is consistent (fully justify your answer): (a) Var(T)= (b) E(T)= and Var(T)= . (c) E(T)=. (d) E(T)= and Var(T)= We were unable to transcribe this imageWe were unable to transcribe this imageoe We were unable to transcribe this imageWe were unable to transcribe this imageWe were unable to transcribe this...
Let be a sample (size n=1) from the exponential distribution, which has the pdf , where is an unknown parameter. Let's define a statistic as . Is a sufficient statistic for ? We were unable to transcribe this imagef(x: λ) = Xe We were unable to transcribe this imageT(X) = 1122 T(X) We were unable to transcribe this image
Suppose X1, X2, ..., Xn are independent and identically distributed (iid) with a Uniform -0,0 distri- bution for some unknown e > 0, i.e., the Xi's have pdf Suppose X1, X2,..., Xn are independent and identically distributed (iid f(3) = S 20, if –0 < x < 0; 20 0, otherwise. (a) (4 pts) Briefly explain why or why not this is an exponential family (b) (5 pts) Find one meaningful sufficient statistic for 0. (By "meaningful”, I mean it...
Suppose constitute a random sample drawn from a population N(, ) and constitute a random sample drawn from another population N(, ). The two samples are drawn independently. Derive a generalised likelihood ratio test for testing against where and are positive constants such that > . We were unable to transcribe this imageWe were unable to transcribe this imageWe were unable to transcribe this imageWe were unable to transcribe this imageμ2 We were unable to transcribe this imageWe were unable...
Independent random samples X1, X2, . . . , Xn are from exponential distribution with pdfs , xi > 0, where λ is fixed but unknown. Let . Here we have a relative large sample size n = 100. (ii) Notice that the population mean here is µ = E(X1) = 1/λ , population variance σ^2 = Var(X1) = 1/λ^2 is unknown. Assume the sample standard deviation s = 10, sample average = 5, construct a 95% large-sample approximate confidence...
Let be a sample (size n = 1) from the exponential distribution, which has the pdf where is an unknown parameter. Let's define a statistic as . Is a sufficient statistic for ? We were unable to transcribe this imagef(x: λ) = Xe We were unable to transcribe this imageT(X) = 11>2 T(X) We were unable to transcribe this image