BEST代写-线上留学生作业代写 & 论文代写专家

成立于2015年的老牌留学生代写品牌-BEST代写提供超过百门冷热门留学学科的作业和论文代写服务。全网BEST原创,高质,准时的留学生代写。

代码代写|STAT 519 Homework 1 Spring 2023

代码代写|STAT 519 Homework 1 Spring 2023

这是一篇来自美国的关于通过代码解决下面一些问题的代写

具体作业要求需要显示计算和/或提供一个解释,以证明您的答案。对于计算部分,使用提供的降价模板文件提交您的代码和图形。

 

Group discussion of homework problems is allowed, but the fifinal write-up should be individual. For each problem please show calculations and/or provide an explanation to justify your answer. For the computing parts, use the provided RMarkdown template fifile to submit your code and fifigures.

Problem 1 (10 points)

Exercise 7.2. For part b, you may use an offff-the-shelf optimization solver, e.g.

R’s optim function.

Problem 2 (5 points)

Exercise 7.4

Problem 3 (10 points)

Exercise 7.6 (b), (c)

Problem 4 (15 points)

Simulate N = 1000 independent random samples of sizes n = 10, 100, 1000, 10000 from the pdf given in Exercise 7.6 with θ = 1.

(a) Plot a separate histogram of the MLEs computed over the 1000 random samples for each of the values of n.

(b) Repeat part (a) for the MOM estimator.

(c) Plot the average squared error for the MLE as a function of n. The average squared error is given by

1/N N Xk=1(θ ˆ k(X1, . . . , Xn) θ) 2 ,

where θ ˆ k(X1, . . . , Xn) is the MLE of the kth random sample of size n.

(d) Repeat part (c) for the MOM estimator.

(e) Brieflfly discuss the similarities/difffferences between the MLE and MOM based on your simulation results. Put your discussion in the RMark

down fifile.

Problem 6 (10 points)

Exercise 7.9

Problem 7 (15 points)

Let X be a continuous random variable with cdf FX and pdf fX such that E|X| < .

  • Prove that E|X a| < for all a.
  • Prove that for all a

E|X a| ≥ E|X m|

where m is the median of X, i.e., FX(m) =12. In other words, the median minimizes the function g(a) = E|X a|.

Problem 8 (15 points)

Let X be a continuous random variable with cdf FX and pdf fX such that E|X| < . Let τ (0, 1) denote a quantile and defifine the “check-loss” function as

ρτ (u) = ((τ 1)u i fu < 0

τu ifu 0.

  • Prove that E[ρτ (X a)] < for all a.
  • Prove that for all a

E[ρτ (X a)] E[ρτ (X qτ )]

where qτ is the τ th quantile of X, i.e., FX(qτ ) = τ . In other words, the τ th quantile minimizes the function g(a) = E[ρτ (X a)].

bestdaixie

评论已关闭。