chapter 6: statistics and sampling distributions · 6.2the distribution of the sample mean 6.3the...
TRANSCRIPT
Chapter 6: Statistics and Sampling Distributions
MATH450
September 7th, 2017
MATH450 Chapter 6: Statistics and Sampling Distributions
Topics
Week 1 · · · · · ·• Chapter 1: Descriptive statistics
Week 2 · · · · · ·• Chapter 6: Statistics and SamplingDistributions
Week 4 · · · · · ·• Chapter 7: Point Estimation
Week 7 · · · · · ·• Chapter 8: Confidence Intervals
Week 10 · · · · · ·• Chapter 9: Test of Hypothesis
Week 13 · · · · · ·• Two-sample inference, ANOVA, regression
MATH450 Chapter 6: Statistics and Sampling Distributions
Overview
6.1 Statistics and their distributions
6.2 The distribution of the sample mean
6.3 The distribution of a linear combination
6.4 Distributions based on a normal random sample
Order: 6.1 → 6.3 → 6.2
Depending on our progress, materials of section 6.4 would beeither
discussed in one lecturedistributed across chapters 8-9-10
MATH450 Chapter 6: Statistics and Sampling Distributions
Random sample
Definition
The random variables X1,X2, ...,Xn are said to form a (simple)random sample of size n if
1 the Xi ’s are independent random variables
2 every Xi has the same probability distribution
MATH450 Chapter 6: Statistics and Sampling Distributions
Recap: Independent random variables
Definition
Two random variables X and Y are said to be independent if forevery pair of x and y values,
P(X = x ,Y = y) = PX (x) · PY (y) if the variables are discrete
or
f (x , y) = fX (x) · fY (y) if the variables are continuous
Property
If X and Y are independent, then for any functions g and h
E [g(X ) · h(Y )] = E [g(X )] · E [h(Y )]
MATH450 Chapter 6: Statistics and Sampling Distributions
Statistics
Definition
A statistic is any quantity whose value can be calculated fromsample data
prior to obtaining data, there is uncertainty as to what valueof any particular statistic will result → a statistic is a randomvariable
the probability distribution of a statistic is sometimes referredto as its sampling distribution
MATH450 Chapter 6: Statistics and Sampling Distributions
Questions for this chapter
Given statistic T computed from sample data X1,X2, . . . ,Xn
Question 1: If we know the distribution of Xi ’s, can we obtainthe distribution of T?Answer: Technically, we can.
Question 2: If we don’t know the distribution of Xi ’s, can westill obtain/approximate the distribution of T?Answer: In a sense, for some statistics
MATH450 Chapter 6: Statistics and Sampling Distributions
Questions for this chapter
Real questions: If T is a linear combination of Xi ’s, can we
compute the distribution of T in some easy cases?
compute the expected value and variance of T?
MATH450 Chapter 6: Statistics and Sampling Distributions
Questions for this section
Real questions: If T = X1 + X2
compute the distribution of T in some easy cases
compute the expected value and variance of T
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 1
Problem
Consider the distribution P
x 40 45 50
p(x) 0.2 0.3 0.5
Let {X1,X2} be a random sample of size 2 from P, andT = X1 + X2.
1 Derive the probability mass function of T
2 Compute the expected value and the standard deviation of T
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 2
Problem
Let {X1,X2} be a random sample of size 2 from the exponentialdistribution with parameter λ
f (x) =
{λe−λx if x ≥ 0
0 if x < 0
and T = X1 + X2.
1 Compute the cumulative density function (cdf) of T
MATH450 Chapter 6: Statistics and Sampling Distributions
Cumulative distribution function
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 2b
Problem
Let {X1,X2} be a random sample of size 2 from the exponentialdistribution with parameter λ
f (x) =
{λe−λx if x ≥ 0
0 if x < 0
and T = X1 + X2.
1 Compute the cumulative density function (cdf) of T
2 Compute the probability density function (pdf) of T
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 2c
Problem
Let X be a random variable with pdf fX (x), and Y = X/2.Compute the pdf fY (y) in term of f .
Reading: Section 4.7
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 2c
Problem
Let {X1,X2} be a random sample of size 2 from the exponentialdistribution with parameter λ
f (x) =
{λe−λx if x ≥ 0
0 if x < 0
and T = X1 + X2.
1 Compute the cumulative density function (cdf) of T
2 Compute the probability density function (pdf) of T
3 Compute the probability density function (pdf) of the mean
X̄ =X1 + X2
2
MATH450 Chapter 6: Statistics and Sampling Distributions
Summary
1 If the distribution and the statistic T is simple, try toconstruct the pmf of the statistic (as in Example 1)
2 If the probability density function fX (x) of X ’s is known, the
try to represent/compute the cumulative distribution (cdf) ofT
P[T ≤ t]
take the derivative of the function (with respect to t )
MATH450 Chapter 6: Statistics and Sampling Distributions
Want to practice more?
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 1*
Problem
Consider the distribution P
x 40 45 50
p(x) 0.2 0.3 0.5
Let {X1,X2} be a random sample of size 2 from P, andT = X1 + X2.
1 Derive the probability mass function of T
2 Compute the expected value and the standard deviation of T
Question: If T = X1 − X2, can you still solve the problem?
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 2*
Problem
Let {X1,X2} be a random sample of size 2 from the exponentialdistribution with parameter λ
f (x) =
{λe−λx if x ≥ 0
0 if x < 0
and T = X1 + X2.
1 Compute the cumulative density function (cdf) of T
2 Compute the probability density function (pdf) of T
Question: If T = X1 + 2X2, can you still solve the problem?Question: If T = X 2
1 + X 22 , can you still solve the problem?
MATH450 Chapter 6: Statistics and Sampling Distributions
Moment generating function
MATH450 Chapter 6: Statistics and Sampling Distributions
Moment generating function
Definition
The moment generating function (mgf) of a continuous randomvariable X is
MX (t) = E (etX ) =
∫ ∞−∞
etx fX (x)dx
Reading: 3.4 and 4.2
MATH450 Chapter 6: Statistics and Sampling Distributions
Moment generating function
Property
Two distributions have the same pdf if and only if they have thesame moment generating function
MATH450 Chapter 6: Statistics and Sampling Distributions
Moment generating function
MATH450 Chapter 6: Statistics and Sampling Distributions
Moment generating function
Definition
Let X1,X2 be a 2 independent random variables and T = X1 + X2,then
MT (t) = MX1(t)MX2(t)
Hint:
MT (t) = E (etT ) = E (et(X1+X2)) = E (etX1 · etX2)
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 3
Problem
Given that the mgf of a Poisson variables with mean λ is
eλ(et−1)
Suppose X and Y are independent Poisson random variables,where X has mean a and Y has mean b. Show that T = X + Yalso follows the Poisson distribution.
MATH450 Chapter 6: Statistics and Sampling Distributions
Example 4
Problem
Given that the mgf of a normal random variables with mean µ andvariance σ2 is
eµt+σ2
2t2
Suppose X and Y are independent normal random variables. Showthat T = X + Y also follows the normal distribution.
MATH450 Chapter 6: Statistics and Sampling Distributions
A bigger picture
MATH450 Chapter 6: Statistics and Sampling Distributions
MATH450 Chapter 6: Statistics and Sampling Distributions
MATH450 Chapter 6: Statistics and Sampling Distributions
Genralization
Inputs: X1,X2, . . . ,Xn
Model: Function f
Output T = f (X1,X2, . . . ,Xn)
Question: If we know the distribution of the input, can we estimatethe distribution of the output?
MATH450 Chapter 6: Statistics and Sampling Distributions
MATH450 Chapter 6: Statistics and Sampling Distributions
Hurricane Irma
MATH450 Chapter 6: Statistics and Sampling Distributions
The forecast process
MATH450 Chapter 6: Statistics and Sampling Distributions