Some Theory Problems (Part II)
Let X1 ~ N(p.a²) and X2 w N(A,O2), where X1 and X2 are independent. Compute
2.i For bivariate random variable (X,Y), derive the shortcut formula for covariance, i.e., show
2.ii If random variables X and Y are independent, what does E(XY] equal?
Definition: Continious random variables X and Y are independent if their joint densify f(x.g) oun be
expressed as the product of their marginal densifies x(x) and fr(g), i.e.,
3.i If X and Y are independent continuous random variables, show that E(XY] = E|X): E(Y]. (Prove
this result using the definition of expectation for continuous random variables and not
by referencing problem 2.ii.)
3.ii Now assume (X,Y) is distributed bivarinte normal. If the linent correlation coeficient P = 0, show X
and Y are independent normal randorn variables.
Let X1 w Binomial(m.p) and X2 ~ Binomial(n.p) and consider the estimator
Prove that P is an unbiased estimator of p.
Lab 4 covers sampling distributions, the central limit theorem (CIT), and maximum likelihood estimation
ench exercise we assume X1,X2, X. is a randorn sample from a Poissom distribution with
rate parameter & 0.
Section I: CLT and Sum of Independent Poisson Random Variables
Set-up Part I: Equivalent Forms of the CLT
Assuming X1,X2 X. is a random sample from some distribution with common mean e and variance
of. then for sufficiently large n:
Expressions (CLT ii) and (CLT. iv) are oftem more desirable because there is mo sample size (n) om the right
hand side of the relation. Recall that the CLT is an asymptotic result (n - 00), hence the sample size
should not appear after taking the limit.
Set-up Part II: Sum of Independent Poisson Random Variables
With some work, one can show that if X1 X2 X2 = Poisson( A). then the sum of X: is also a Poisson
randorn variable with rate nd. i.e.,
Suppose that the number of customers arriving at a gas station (X;) in any given day has a Poisson
distribution with rate A = 100 [customers per day|- The manager of the gas statiom is interested in the
total number of customers (T. = Li-1 X1) arriving in one week and one month Assume that one
weck has n = 7 days and ome month has n = 31 days.
Solve the following problems: (1)-(2)
1. Compute the probability (and approximate probability) that more than 670 customers visit the gas
station in a weck. Solve this problem using the Poisson distribution and the CLT. Compare the
approximate answer to the correct Poissom probability. (Note that we are assuming n = 7 < 30
for the CLT, which is typically discouraged)
2. Compute the probability (and approximate probability) that more than 3000 customers visit the gas
station in a month. Solve this problem using the Poisson distribution and the CLT. Compare the
approximate answer to the correct Poisson probability.
Note the following:
Section II: MLE of Poisson Rate Parameter
Let X1, X2, X. be a random sample from a Poisson distribution with rate parameter A> 0
Solve the following problems: (3)-(6)
3. Derive the maximum likelihood estimator of A. Denote the MLE by is
4. Show that & is an unbiased estimator of &
5. Derive an expression for the standard error of is
6. Identify the large sample distribution of &
This material may consist of step-by-step explanations on how to solve a problem or examples of proper writing, including the use of citations, references, bibliographies, and formatting. This material is made available for the sole purpose of studying and learning - misuse is strictly forbidden.