MAS223 Statistical Inference and Modelling Exercises

Transcription

1 MAS223 Statistical Inference and Modelling Exercises The exercises are grouped into sections, corresponding to chapters of the lecture notes Within each section exercises are divided into warm-up questions, ordinary questions, and challenge questions Note that there are no exercises accompanying Chapter 8 The vast majority of exercises are ordinary questions Ordinary questions will be used in homeworks and tutorials; they cover the material content of the course Warm-up questions are typically easier, often nothing more than revision of relevant material from first year courses Challenge questions are typically harder and test ingenuity Contents Univariate Distribution Theory 2 2 Standard Univariate Distributions 5 3 Transformations of Univariate Random Variables 7 4 Multivariate Distribution Theory 9 5 Transformations of Multivariate Distributions 6 Covariance Matrices and and Multivariate Normal Distributions 4 7 Likelihood and Maximum Likelihood 7

2 Univariate Distribution Theory Warm-up Questions Let X be a random variable taking values in, 2, 3}, with P[X = ] = P[X = 2] = 04 Find P[X = 3], and calculate both E[X] and Var[X] 2 Let Y be a random variable with probability density function (pdf) f(y) given by y/2 for 0 y < 2; f(y) = Find the probability that Y is between 2 and Calculate E[Y ] and Var[Y ] Ordinary Questions 3 Define F : R [0, ] by 0 for y 0; F (y) = y 2 for y (0, ); for y (a) Sketch the function F, and check that it is a distribution function (b) If Y is a random variable with distribution function F, calculate the pdf of Y 4 Let X be a discrete random variable, taking values in 0,, 2}, where P[X = n] = 3 for n 0,, 2} Sketch the distribution function F X : R R 5 Define f : R [0, ] by f(x) = 0 for x < 0; e x for x 0 (a) Show that f is a probability density function (b) Find the corresponding distribution function and evaluate P[ < X < 2] 6 Sketch graphs of each of the following two functions, and explain why each of them is not a distribution function 0 for x 0; (a) F (x) = x for x > 0 0 for x < 0; (b) F (x) = x + sin 2πx for 0 x < ; 4 for x 2

3 7 Let k R and define f : R R by k(x x 2 ) for x (0, ), f(x) = Find the value of k for which f(x) is a probability density function, and calculate the probability that X is greater than 2 8 The probability density function f(x) is given by + x for x 0; f(x) = x for 0 < x < ; 9 Let Find the corresponding distribution function F (x) for all real x F (x) = ex + e x for all real x (a) Show that F is a distribution function, and find the corresponding pdf f (b) Show that f( x) = f(x) (c) If X is a random variable with this distribution, evaluate P[ X > 2] 0 Show that f(x) = π, defined for all x R, is a probability density function +x 2 (a) For which values of r [0, ) is x r dx finite? (b) Show that x 2 if x > f(x) = 0 otherwise is a probability density function (c) Show that the expectation of a random variable X, with the probability density function f given in (b), is not defined (d) Give an example of a random variable Y for which E[Y ] < but E[Y 2 ] is not defined 2 The discrete random variable X has the probability function P[X = x] = (a) Use the partial fractions of x(x+) x(x+) for x N to show that P[X x] =, for all x N x+ (b) Write down the distribution function F (x) of X, for x R Sketch its graph What are the values of F (2) and F ( 3 2 )? (c) Evaluate P[0 X 20] (d) Is E[X] defined? If so, what is E[X]? If not, why not? 3

4 Challenge Questions 3 Show that there is no random variable X, with range N, such that P[X = n] is constant for all n N 4 Recall the meaning of inscribing a cube within a sphere: the cube sits inside of the sphere, with each vertex of the cube positioned on the surface of the sphere It is not especially easy to illustrate this on a two dimensional page, but here is an attempt: Suppose that ten percent of the surface of the sphere is coloured blue, and the rest of the surface is coloured red Show that, regardless of which parts are coloured blue, it is always possible to inscribe a cube within the sphere in such a way as all vertices of the cube are red Hint: A cube has eight corners Suppose that position of the cube is sampled uniformly from the set of possible positions What is the expected number of corners that are red? 4

5 2 Standard Univariate Distributions Warm-up Questions 2 (a) A standard fair dice is rolled 5 times Let X be the number of sixes rolled Which distribution (and which parameters) would you use to model X? (b) A fair coin is flipped until the first head is shown Let X be the total number of flips, including the final flip on which the first head appears Which distribution (and which parameter) would you use to model X? Ordinary Questions 22 Let λ > 0 Write down the pdf f of the random variable X, where X Exp(λ), and calculate its distribution function F Hence, show that is constant for t > 0 f(t) F (t) 23 Let λ > 0 and let X be a random variable with Exp(λ) distribution Let Z = X, that is let Z be X rounded down to the nearest integer Show that Z is geometrically distributed with parameter p = e λ 24 Let µ R Let X and X 2 be independent random variables with distributions N(µ, ) and N(µ, 4), respectively Let T, T 2 and T 3 be defined by T = X + X 2 2, T 2 = 2X X 2, T 3 = 4X + X 2 5 Find the mean and variance of T, T 2 and T 3 Which of E[T ], E[T 2 ] and E[T 3 ] would you prefer to use as an estimator of µ? 25 Let X be a random variable with Ga(α, β) distribution (a) Let k N Show that E[X k ] = α(α + ) (α + k ) β k Hence, calculate µ = E[X] and σ 2 = Var(X) and verify that these formulas match the ones given in lectures [ ( (b) Show that E X µ ) 3 ] = 2 σ α 26 (a) Using R, you can obtain a plot of, for example, the pdf of a Ga(3, 2) random variable between 0 and 0 with the command curve(dgamma(x,shape=3,scale=2),from=0,to=0) Use R to investigate how the shape of the pdf of a Gamma distribution varies with the different parameter values In particular, fix a value of β, see how the shape changes as you vary α 5

6 (b) Investigate the effect that changing parameters values has on the shape of the pdf of the Beta distribution To produce, for example, a plot of the pdf of Be(4, 5), use curve(dbeta(x,shape=4,shape2=5),from=-,to=2) 27 Suggest which standard discrete distributions (or combination of them) we should use to model the following situations (a) Organisms, independently, possess a given characteristic with probability p A sample of k organisms with the characteristic is required How many organisms will need to be tested to achieve this sample? (b) In Texas Hold em Poker, players make the best hand they can by combining two cards in their hand with five community cards that are placed face up on the table At the start of the game, a player can only see their own hand The community cards are then turned over, one by one A player has two hearts in her hand Three of the community cards have been turned over, and only one of them is a heart How many hearts will appear in the remaining two community cards? Use a computer to find the probability of seeing k = 0,, 2 hearts 28 Let X be a N(0, ) random variable Use integration by parts to show that E[X n+2 ] = (n + )E[X n ] for any n = 0,, 2, Hence, show that E[X n 0 if n is odd; ] = ()(3)(5) (n ) if n is even 29 Let X N(µ, σ 2 ) Show that E[e X σ2 µ+ ] = e 2 Challenge Questions 20 Let X be a random variable with a continuous distribution, and a strictly increasing distribution function F Show that F (X) has a uniform distribution on (0, ) Suggest how we might use this result to simulate samples from standard distributions 2 Prove that Γ( 2 ) = π 6

7 3 Transformations of Univariate Random Variables Warm-up Questions 3 Let g : R R be given by g(x) = x 3 + (a) Sketch the graph of g, show that g is strictly increasing, and find its inverse function g (b) Let R = [0, 2] Find g(r) Ordinary Questions 32 Let X be a random variable with pdf x 2 for x > f X (x) = Define g(x) = e x and let Y = g(x) (a) Show that g(x) is strictly increasing Find its inverse function g (y), and dg (y) dy (b) Identify the set R X on which f X (x) > 0 Sketch g and show that g(r X ) = (e, ) (c) Deduce from (a) and (b) that Y has pdf (log y) 2 for y > e y f Y (y) = 33 Let X be a random variable with the uniform distribution on (0, ), and let Y = log X λ where λ > 0 Show that Y has an Exp(λ) distribution 34 Let α, β > 0 (a) Show that B(α, β) = B(β, α) (b) Let X be a random variable with the Be(α, β) distribution Show that Y = X has the Be(β, α) distribution 35 Let α > 0 (a) Show that B(α, ) = α (b) Let X Be(α, ) distribution Let Y = r X for some positive integer r Show that Y also has a Beta distribution, and find its parameters 36 Let X have a uniform distribution on [, ] Find the pdf of X and identify the distribution of X 37 Let α, β > 0 and let X Be(α, β) Let c > 0 and set Y = c/x Find the pdf of Y 7

8 38 Let Θ be an angle chosen according to a uniform distribution on ( π, π ), and let 2 2 X = tan Θ Show that X has the Cauchy distribution 39 Let X be a random variable with the pdf + x for < x < 0; f(x) = x for 0 < x < ; Find the probability density functions of (a) Y = 5X + 3 (b) Z = X 30 Let X have the uniform distribution on [a, b] (a) For [a, b] = [, ], find the pdf of Y = X 2 (b) For [a, b] = [, 2], find the pdf of Y = X 3 Let X have a uniform distribution on [, ] and define g : R R by 0 for x 0; g(x) = x 2 for x > 0 Find the distribution function of g(x) 32 Let X be a random variable with the Cauchy distribution Show that X also has the Cauchy distribution Challenge Questions 33 If we were to pretend that g(x) = /x was strictly monotone, we could (incorrectly) apply Lemma 3 and use the formula f Y (y) = f X (g (y)) dg to solve 32 We dy would still arrive at the correct answer Can you explain why? Can you construct another example of a case in which the relationship f Y (y) = f X (g (y)) dg holds, but where the function g is not monotone? dy 34 Let Y and α, β be as in Question 37 (a) If α >, show that E[Y ] = c(α+β ) α (b) If α show that E[Y ] is not defined 8

9 4 Multivariate Distribution Theory Warm-up questions 4 Let T = (x, y) : 0 < x < y} Define f : R 2 R by e 2x y for (x, y) T ; f(x, y) = Sketch the region T Calculate 0 that they are equal 42 Sketch the following regions of R 2 (a) S = (x, y) : x [0, ], y [0, ]} (b) T = (x, y) : 0 < x < y} (c) U = (x, y) : x [0, ], y [0, ], 2y > x} f(x, y) dx dy and f(x, y) dy dx, and verify Ordinary Questions 43 Let (X, Y ) be a random vector with joint probability density function ke (x+y) if 0 < y < x f X,Y (x, y) = (a) Using that P[(X, Y ) R 2 ] =, find the value of k (b) For each of the regions S, T, U in 42, calculate the probability that (X, Y ) is inside the given region (c) Find the marginal pdf of Y, and hence identify the distribution of Y 44 Let S = [0, ] [0, ], and let U and V have joint probability density function 4u+2v (u, v) S; 3 f U,V (u, v) = (a) Find P[U + V ] (b) Find P[V U 2 ] 45 For the random variables U and V in Exercise 44: (a) Find the marginal pdf f U (u) of U (b) Find the marginal pdf f V (v) of V (c) For v such that f V (v) > 0, find the conditional pdf f U V =v (u) of U given V = v (d) Check that each of f U, f V and f U V =v integrate over R to (e) Calculate the two forms of conditional expectation, E[U V = v], and E[U V ] 9

10 46 Let (X, Y ) be a random vector with joint probability density function y x x [, 0], y [0, ]; 2 x+y f X,Y (x, y) = x [0, ], y [0, ]; 2 Find the marginal pdf of X Show that the correlation coefficient X and Y is zero Show also that X and Y are not independent 47 Let X be a random variable Let Z be a random variable, independent of X, such that P[Z = ] = P[Z = ] = Let Y = XZ 2 (a) Show that X and Y are uncorrelated (b) Give an example in which X and Y are not independent 48 Let X and Y be independent random variables, with 0 < Var(X) = Var(Y ) < Let U = X + Y and V = XY Show that U and V are uncorrelated if and only if E[X] + E[Y ] = 0 49 Let λ > 0 Let X have an Exp(λ) distribution, and conditionally given X = x let U have a uniform distribution on [0, x] Calculate E[U] and Var(U) 40 Let k R and let (X, Y ) have joint probability density function kx sin(xy) for x (0, ), y (0, π), f X,Y (x, y) = (a) Find the value of k (b) For x (0, ), find the conditional probability density function of Y given X = x (c) Find E[Y X] 4 Let U have a uniform distribution on (0, ), and conditionally given U = u let X have a uniform distribution on (0, u) (a) Find the joint pdf of (X, U) and the marginal pdf of X (b) Show that E[U X = x] = x log x 42 Let (X, Y ) have a bivariate distribution with joint pdf f X,Y (x, y) Let y 0 R be such that f Y (y 0 ) > 0 Show that f X Y =y0 (x) is a probability density function Challenge Questions 43 Give an example of random variables (X, Y, Z) such that P[X < Y ] = P[Y < Z] = P[Z < X] = 2 3 0

11 5 Transformations of Multivariate Distributions Warm-up Questions 5 (a) Define u = x and v = 2y Sketch the images of the regions S, T and U from question 42 in the (u, v) plane (b) Define u = x + y and v = x y Sketch the images of the regions S, T and U from question 42 in the (u, v) plane Ordinary Questions 52 The random variables X and Y have joint pdf given by xe y if x (0, 2), y (0, ) f X,Y (x, y) = Define u = u(x, y) = x + y and v = v(x, y) = 2y Let U = u(x, Y ) and V = v(x, Y ) (a) Find the inverse transformation ) x = x(u, v) and y = y(u, v) and calculate the value of J = det ( x/ u x/ v y/ u y/ v (b) Sketch the set of (x, y) for which f X,Y (x, y) is non-zero Find the image of this set in the (u, v) plane (c) Deduce from (a) and (b) that the joint pdf of (U, V ) is (u v 2 f U,V (u, v) = )e v/2 for v > 0, u (2v, 2v + 4) 53 The random variables X and Y have joint pdf given by 2 f X,Y (x, y) = (x + y)e (x+y) for x, y 0 0 elsewhere Let U = X + Y and W = X (a) Find the joint pdf of (U, W ) and the marginal pdf of U (b) Recognize U as a standard distribution and, using the result of Question 25(a), evaluate E[(X + Y ) 5 ] 54 Let X and Y be a pair of independent random variables, both with the standard normal distribution Show that the joint pdf of (U, V ) where U = X 2 and V = X 2 + Y 2 is given by 8π f U,V (u, v) = e v/2 u /2 (v u) /2 for 0 u v

12 55 Let (X, Y ) be a random vector with joint pdf 2e (x+y) x > y > 0; f X,Y (x, y) = (a) If U = X Y and V = Y/2, find the joint pdf of (U, V ) (b) Show that U and V are independent, and recognize their (marginal) distributions as standard distributions 56 Let X and Y be a pair of independent and identically distributed random variables Let U = X + Y and V = X Y (a) Show that Cov(U, V ) = 0, and give an example to show that U and V are not necessarily independent (b) Show that U and V are independent in the special case where X and Y are standard normals 57 Let X and Y be independent random variables with distributions Ga(α, β) and Ga(α 2, β) respectively Show that the random variables U = X and V = X + Y are X+Y independent with distributions Be(α, α 2 ) and Ga(α + α 2, β) respectively 58 As part of Question 57, we showed that if X and Y are independent random variables with X Ga(α, β) and Y Ga(α 2, β), then X + Y Ga(α + α 2, β) (a) Use induction to show that for n 2, if X, X 2,, X n are independent random variables with X i Ga(α i, β) then ( n n ) X i Ga α i, β i= (b) Hence show that for n, if Z, Z 2,, Z n are independent standard normal random variables then n Zi 2 χ 2 n i= You may use the result of Example 2, which showed that this was true in the case n = (and, recall that the χ 2 distribution is a special case of the Gamma distribution) 59 Let X and Y be a pair of independent random variables, both with the standard normal distribution Show that U = X/Y has the Cauchy distribution, with pdf i= for all u R f U (u) = π + u 2 2

13 50 Let n N The t distribution (often known as Student s t distribution) is the univariate random variable X with pdf f X (x) = Γ( n+ 2 ) nπγ( n 2 ) ) n+ ( + x2 2, n for all x R Here, n is a parameter, known as the number of degrees of freedom Let Z be a standard normal random variable and let W be a chi-squared random variable with n degrees of freedom, where Z and W are independent Show that X = Z W/n has the t distribution with n degrees of freedom Challenge Questions 5 The formula (52), from the typed lecture notes, holds whenever the transformation u = u(x, y), v = v(x, y) is both one-to-one and onto, and for which the Jacobian matrix of derivatives exists Use this fact to provide an alternative proof of Lemma 3 (ie of the univariate transformation formula) 52 Let X Exp(λ ) and Y Exp(λ 2 ) be independent Show that U = min(x, Y ) has distribution Exp(λ + λ 2 ), and that P[min(X, Y ) = X] = λ λ +λ 2 Extend this result, by induction, to handle the minimum of finitely many exponential random variables Let W = max(x, Y ) Show that U and W U are independent 3

14 6 Covariance Matrices and and Multivariate Normal Distributions Warm-up Questions 6 Let A = ( ) 2 and Σ = ( ) 4 9 (a) Calculate det(a) and find both AΣ and AΣA T (b) Let A be the vector (2, ) Show that AΣA T = 2 Ordinary Questions 62 Let X = (X, Y ) T be a random vector with ( ) 0 E[X] =, Cov(X) = Let U = X + Y and V = 2X 2Y + ( ) 2 (a) Write down a square matrix A and a vector b R 2 such that U = AX + b (b) Show that the mean vector and covariance matrix of U are given by ( ) ( ) 5 2 E[U] =, Cov(U) = 2 4 (c) Show that the correlation coefficient ρ of U and V is equal to 5 63 Let X and Y be independent standard normal random variables (a) Write down the covariance matrix of the random vector X = (X, Y ) T (b) Let R be the rotation matrix ( ) cos θ sin θ sin θ cos θ Show that RX has the same covariance matrix as X 64 Three (univariate) random variables X, Y and Z have means 3, 4 and 6 respectively and variances, and 25 respectively Further, X and Y are uncorrelated; the correlation coefficient between X and Z is 5 and that between Y and Z is 5 Let U = X + Y Z and W = 2X + Z 4 and set U = (U, W ) T (a) Find the mean vector and covariance matrix of X = (X, Y, Z) T (b) Write down a matrix A and a vector b such that U = AX + b (c) Find the mean vector and covariance matrix of U 4

15 (d) Evaluate E[(2X + Z 6) 2 ] 65 Suppose that the random vector X = (X, X 2 ) T follows the bivariate normal distribution with E[X ] = E[X 2 ] = 0, Var(X ) =, Cov(X, X 2 ) = 2 and Var(X 2 ) = 5 (a) Calculate the correlation coefficient of X and X 2 Are X and X 2 independent? (b) Find the mean and the covariance matrices of Y = ( 2 ) X and Z = What are the distributions of Y and Z? ( ) 2 X Let X and X 2 be bivariate normally distributed random variables each with mean 0 and variance, and with correlation coefficient ρ (a) By integrating out the variable x 2 in the joint pdf, verify that the marginal distribution of X is indeed that of a standard univariate normal random variable Hint: Use the fact that the integral of a N(µ, σ 2 ) pdf is equal to (b) Show, using the usual formula for the conditional pdf that the conditional pdf of X 2 given X = x is N(ρx, ρ 2 ) 67 Let X = (X, X 2 ) T have( a N 2 (µ, ) Σ) distribution with mean vector µ = (µ, µ 2 ) T and σ 2 covariance matrix Σ = σ 2 Let σ 2 σ 2 2 A = ( σ ) σ 2 σ 2 σ Find the distribution of Y = AX, and deduce that any bivariate normal random vector can be transformed by a linear transformation into a vector of independent normal random variables 68 The random vector X = (X, X 2, X 3 ) T has an N 3 (µ, Σ) distribution where µ = (,, 2) T and Σ = (a) Find the correlation coefficients between X and X 2, between X and X 3 and between X 2 and X 3 (b) Let Y = X + X 3 and Y 2 = X 2 X Find the distribution of Y = (Y, Y 2 ) T and hence find the correlation coefficient between Y and Y 2 69 (a) Let X be a (univariate) standard normal random variable and let Y = X Does the random vector X = (X, Y ) T have a bivariate normal distribution? (b) Let X and Y be two independent (univariate) standard normal random variables, and let Z be a random variable such that P[Z = ] = P[Z = ] = Are 2 XZ and Y Z independent, and does the random vector X = (XZ, Y Z) T have a bivariate normal distribution? 5

16 Challenge Questions 60 Recall that an orthogonal matrix R is one for which R = R T, and recall that if an n n matrix R is orthogonal, x is an n-dimensional vector, and y = Rx then n i= y2 i = y y = x x = n i= x2 i Let X = (X, X 2,, X n ) be a vector of independent normal random variables with common mean 0 and variance σ 2 Let R be an orthogonal matrix and let Y = RX (a) Show that Y is also a vector of independent normal random variables, with common mean 0 and variance σ 2 (b) Suppose that all the elements in the first row of R are equal to n (you may assume that an orthogonal matrix exists with this property) Show that Y = n X, where X is the sample mean of X, and that n i=2 Y 2 i = n Xi 2 n X 2 (c) Hence, use Question 58 to deduce that, if s 2 = sample variance of X, (n )s 2 χ 2 n, and that it is independent of X σ 2 i= n ( n i= X2 i n X 2) is the (d) Let µ R Deduce that the result of part (c) also holds if the X i have mean µ (so as they are iid N(µ, σ 2 ) random variables) 6 Let Z, Z 2, be independent identically distributed normal random variables with mean µ and variance σ 2 We regard n samples of these as a set of data Write Z and s 2 respectively for the sample mean and variance Combine 60(d) and 50 to show that the statistic n( Z µ) X = s has the t distribution with n degrees of freedom 6

17 7 Likelihood and Maximum Likelihood Warm-up Questions 7 Let f : R R by f(θ) = e θ2 +4θ (a) Find the first derivative of f, and hence identify its turning point(s) (b) Calculate the value of the second derivative of f at these turning point(s) Hence, deduce if the turning point(s) are local maxima or local minima ( n ) 72 Let (a i ) n i= be a sequence with a i (0, ) Show that log a i = n log a i i= i= Ordinary Questions 73 A sample of 3 is obtained from a geometric distribution X with unknown parameter θ That is, P[X = x] = θ x ( θ) for x 0,, 2, } (a) Given this sample, find the likelihood function L(θ; 3), and state its domain Θ (b) Find the maximum likelihood estimate of θ 74 A sample of 4 is obtained from a Bi(n, θ) distribution (a) Assuming that n is known to be 9 and that θ is unknown, write down the likelihood function of θ, for θ [0, ], given the (single) data point 4 Use a software package of your choice to plot a graph of it (b) Find the maximum likelihood estimate of θ, given this data point 75 A sample (x, x 2, x 3 ) of three observations from a Poisson distribution with parameter λ, where λ is known to be in Λ =, 2, 3}, gives the values x = 4, x 2 = 0, x 3 = 3 Find the likelihood of each of the possible values of λ, and hence find the maximum likelihood estimate 76 Given the set of iid samples x = (x, x 2,, x n ), for some n > 0, write down the likelihood function L(θ; x), in each of the following cases In each case you should give both the function, in simplified form where possible, and the parameter set Θ The data are iid samples from: (a) the exponential distribution Exp(λ) with λ = /θ (b) the binomial Bi(m, θ) distribution, where m is known (c) the normal N(µ, θ), where µ is known (d) the gamma distribution Ga(θ, 4) (e) the beta distribution Be(θ, θ) 77 For each of 76(a)-(e), find the log likelihood, and simplify it as much as you can 7

18 78 The IGa(α, β) distribution is the distribution of /U if U Ga(α, β) Repeat 76 in the case where the data are iid samples from the inverted gamma distribution IGa(, θ) Find and simplify the corresponding log likelihood 79 A set of iid samples x = (x, x 2,, x n ) T is taken from a geometric distribution with unknown parameter θ, so that the probability function of X i is p(x i ) = P[X i = x i ] = ( θ) x i θ, x i = 0,, 2,, 0 < θ < Find the maximum likelihood estimate of θ based on the above sample 70 Find the maximum likelihood estimate of θ when the data x = (x, x 2,, x n ) are iid samples from the binomial Bi(m, θ) distribution, where m is known, as in question 76/77(b) 7 Find the maximum likelihood estimate of θ when the data x = (x, x 2,, x n ) are iid samples from the normal N(µ, θ), where µ is known, as in question 76/77(c) Show that this estimator is unbiased (ie that E[ˆθ] = σ 2 ) 72 Find the maximum likelihood estimate of θ (0, ) when the data x = (x, x 2,, x n ) are iid samples from the gamma distribution Ga(3, θ) If x = n n i= x i = 3, calculate the maximum likelihood estimate ˆθ and show that it does not depend on the sample size n 73 Suppose that a set of iid samples x = (x,, x n ) is taken from a negative binomial distribution, so that each X i has probability function ( ) xi + r p(x i ) = P (X i = x i ) = θ r ( θ) x i r for some success probability θ satisfying 0 θ, where x i = 0,, 2, and r is the total number of successes If r is known, find the maximum likelihood estimate of θ 74 As in question 74, an observation from a Bi(n, θ) gives the value x = 4 (a) If θ is known to be 3 but n is unknown, with possible range n 4, 5, 6,, }, 4 write down a formula for the likelihood function of n and calculate its values for n = 4, 5, 6 and 7 (b) Find the maximum likelihood estimator of n 75 Suppose we have data x = (x,, x n ), which are iid samples from a N(µ, σ 2 ) distribution, where µ is unknown and σ 2 is known (a) Find the log-likelihood function of µ (b) Show that the maximum likelihood estimator of µ is ˆµ = n (c) Let k (0, ) Find the k-likelihood region for µ n x i i= 8

19 76 As in question 74(a) an observation from a Bi(n, θ) distribution with n = 9 gives the value x = 4 (a) Find the range of values for which the log likelihood is within 2 of its maximum value [You may do this either by inspection of a plot, or by using a computer package to solve the inequality numerically] (b) A traditional approximate 95% confidence interval here would be of the form ˆθ ± 96 ˆθ( ˆθ), where ˆθ = x/n Compare your answer to (a) to what this would n give (c) Repeat this analysis with n = 90 and x = 40, and comment on your results 77 A set of iid samples x = (x, x 2,, x n ) T is taken from an inverse Gaussian distribution (known also as the Wald distribution), with pdf ) θ f(x) = ( 2πx exp θ(x µ)2 when x > 0, 3 2µ 2 x and f(x) = x for x 0, with parameters µ, θ > 0 (a) Assuming that µ is known, find the maximum likelihood estimate of θ based on the above sample (b) If both µ and θ are unknown, find the maximum likelihood estimate of (µ, θ) based on the sample given Challenge Questions 78 The Pareto distribution has parameters α > 0 and β > 0, and pdf f(x; θ) = αβα x α+ when x β and f(x; θ) = 0 when x < β A set of n iid samples x = (x, x 2,, x n ) T are taken from a Pareto distribution, where the parameters θ = (α, β) are both unknown Find the maximum likelihood estimate of θ based on the above sample 9