Chapter 7
Chapter 7
Chapter 7
Pilani Campus
4
BITS Pilani, Pilani Campus
Desirable Properties
1. ˆ to be unbiased for .
2. ˆ to have a small variance for large sample size.
Unbiased estimator:
An estimator ˆ is an unbiased estimator for a
population parameter if and only if
E ˆ = .
5 BITS Pilani, Pilani Campus
Sample Mean
Ex.7.1. Let X 1 , X 2 , , X n be a random sample of size n from a distribution
with mean . The sample mean, X , is an unbiased estimator for .
Sol. Here X 1 , X 2 , , X n is a random sample of size n from a distribution
with mean . Then, the sample mean, X , is
1
X X1 X 2 X n
n
E X E X1 X 2 X n
1
n
1
= E X 1 +E X 2 +E X 3 + +E X n
n
Thus, the sample mean, X , is an unbiased estimator for .
6 BITS Pilani, Pilani Campus
Variance of Sample Mean
2
Ex.7.2. Show that Var X
2
X
n
x 1e ; x 0, 0, 0
f x; ,
0 ; otherwise
14 BITS Pilani, Pilani Campus
Maximum Likelihood Estimation
m
nm xi
m
xi
i 1 0
L '( p) i 1
L( p ) p 1 p
m
m
nm xi p xi (1 p )
i 1 i 1
m
Xi X
Step 3 : The estimator of p is then obtained as pˆ i 1
nm n
Why does this estimator maximize likelihood function?
17 BITS Pilani, Pilani Campus
Examples: MLE
Ex.7.6. Use MLE to estimate parameters of exponential distribution
x
1
f x; e ; x 0, 0
Sol.
Step 1 : The log likelihood function for exponential distribution is
n
xi
ln L | x ln L ; x1 , x2 ,..., xn n ln
i 1
Step 2 : The corresponding log likelihood equation is
ln L 0
1 n
Step 3 : The estimator of is then obtained as ˆ X i
n i 1
18 BITS Pilani, Pilani Campus
Examples: MLE
HW.7.7. Use MLE to estimate the parameter of Poisson distribution.
X X
n
2
i
n 1 2
Note that the ML estimator for is ˆ 2 2 i 1
S .
n n
Thus M-L estimator for 2 is not unbiased.
Sol.
Step 1 : The log likelihood function is
ln L | t ln L , ; t1 , t2 ,..., tn
n ln n ln 1 ln ti i
n n
t
i 1 i 1
Step 2 : The corresponding log likelihood equations are
ln L 0 and ln L 0
22 BITS Pilani, Pilani Campus
Examples: MLE
This gives
1 n
ln L , ; t1 , t2 ,..., tn 0 ti 0
n i 1
n n ti ti
ln L , ; t1 , t2 ,..., tn 0 1 ln 0
i 1
Step 3 : The estimates of and are then obtained from
n
1 1 n i ln ti
t
1 n
1
ln ti i 1
0 and ti
n i 1 n
n i 1
i
t
i 1
i 1
i 1
Theorem 7.3.3:
Let X be a random variable with moment
generating function mx(t). Let Y = α + βX. The
moment generating function for Y is
mY(t) = e αt mx(βt)
32
BITS Pilani, Pilani Campus
HW 7.15 (Q 44: Distribution of a sum of
independent chi-squared random variables)
Let Y=X1+X2…+Xn .
Show that Y is a chi-squared random variable with
degrees of freedom where γ= ∑ γi
33
BITS Pilani, Pilani Campus
Homework
34
BITS Pilani, Pilani Campus
Homework
35
BITS Pilani, Pilani Campus
BITS Pilani
Pilani Campus
P X z 2 X z 2 1
n n
Hence, the confidence interval for population mean having confidence
level 100 1 % is given as X z 2 , X z 2 .
n n
The endpoints of the confidence interval is called confidence limits.
Hence, 95% CI for is given as X 1.96 , X 1.96 .
n n
That is, P X 1.96 X 1.96 0.95
n n
Sol. (b)
Step 1: Here n 50, x 15.68, =3.27, and 0.05. We need CI for .
Step 2: As 0.05, we need to find z 2 such that P Z z 2 =0.975.
From cumulative normal distribution table, we see z 2 1.96.
Step 3: The CI for known is x z0.025 , x z0.025 14.77,16.59
n n
HW7.23. A certain group of welfare recipients receives pension benefits of Rs. 45000
per month with a standard deviation of Rs. 7500. If a random sample of 25 people is
taken, what is the probability that their mean pension benefit will be greater than
Rs. 47000 or less than Rs. 43000 per month ?