
15. Suppose that X ~ B(mp). Show that, for any positive integer k, the parameter θ...
For any two positive integers a, b, define k(a,b) to be the largest k such that a* | b but ak+1b. Given two positive integers x, y, show that (a) k(a, gcd(x, y)) = min{k(a, x), k(a, y)} for any positive integer a (b) k(a, lcm(z, y)) = max{k(a,a),k(a, y)} for any positive integer a. Hint: Think of the prime factorization of the numbers
For any two positive integers a, b, define k(a,b) to be the largest k such that...
5. For X follows Exp(6) (exponential distribution with parameter θ), a hypothesis test rejects the null hypothesis Ho : θ-1 when X k versus H1 : θ > 1. (a) Show that for any k greater than -log(0.05), the test has the probability of type I error less than 0.05 (b) Show that the power of the test at θ-10 is larger when k-1 than k-2. (c) Let k-_ log(0.05), calculate the power function in terms of θ when θ...
1. a) Deduce from the relation Fk+1F-1-F2 = (-1)k, that for any positive integer k > 2 Fk b) Deduce from part a) that for any positive integer n, Fn+1 = Σ 1)k+1
1. a) Deduce from the relation Fk+1F-1-F2 = (-1)k, that for any positive integer k > 2 Fk b) Deduce from part a) that for any positive integer n, Fn+1 = Σ 1)k+1
We have n independent observations from a geometric distribution with unknown parameter θ. Po(X,-k-θ(1-0)4-1 for k-1, 2, 3, . . . We wish to test the null hypothesis θ-1/2 versus the alternative θ 7|/2. we can show that the MLE θ-1/2. Write out the appropriate LRT statistic as a function of the r, the mean of the observations
1. (15 points) For any parameter t, show that the R-K method (ks = /(z,-+ (1-1)h,y,, + (1-1) ). has the local truncation error O(h3)
1. (15 points) For any parameter t, show that the R-K method (ks = /(z,-+ (1-1)h,y,, + (1-1) ). has the local truncation error O(h3)
Question 3: A random variable X has a Bernoulli distribution with parameter θ є (0,1) if X {0,1} and P(X-1)-θ. Suppose that we have nd random variables y, x, following a Bernoulli(0) distribution and observed values y1,... . Jn a) Show that EIX) θ and Var[X] θ(1-0). b) Let θ = ỹ = (yit . .-+ yn)/n. Show that θ is unbiased for θ and compute its variance. c) Let θ-(yit . . . +yn + 1)/(n + 2) (this...
12. Suppose XIX, iid X, P(θ, l), where P(0,1) is the one-parameter Pareto distribution with density f(x)-0/10+1 for l < x < 00, Assume that θ >2, so that the model θ/(0-1)(8-2)2 (a) obtain the MME θι from the first moment equation and the MIE θ2 (b) Obtain the asymptotic distributions of these two estimators. (c) Show that the ML is asymptotically superior to the MME P(0,1) has finite mean θ/(9 -1 ) and variance
3.11 Theorem. Suppose f(x)-a"x" + an-lx"-+ + ao is a poly- nomial of degree n > 0 and suppose an > 0. Then there is an integer k such that ifx >k, then f(x)> 0. Note: We are only assuming that the leading coefficient an is greater than zero. The other coefficients may be positive or negative or zero. The next theorem extends the idea that polynomials get positive and roughly states that not only do they get positive, but...
Problem 4. Let w be a positive continuous function and let n be a nonnegative integer. Equip P.(R) with the inner product (p,q) = $' p(x)q(x)"(x) dx. You do not need to check that this is an inner product. (a) Prove that P.(R) has an orthonormal basis po..., Pr such that deg pk = k for each k. (b) Show that (Pk, pk) = 0 for each k, where the polynomials pį are from the preceding part. Here pé denotes...
3. Suppose that X is a nonegative integer valued random variable. Show that E[X] = P(X ). Hint: Start with the formula EX= k= k, Now try to rearrange the terms. P(X = k) and for all positive integers k write