Download 12.1 Estimation

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
12.1 Estimation of the Difference between the Means
Motivating example:
Objective:
we want to determine the difference between the mean ages of the
customers shopping at the inner-city and the suburban.
1 : the mean age of all customers who shop at the inner-city store
 2 : the mean age of all customers who shop at the suburban store
 12 : the variance of the ages of all customers who shop at the inner-city store
 22 : the variance of the ages of all customers who shop at the suburban store
We want to estimate the difference
1  2 .
Let
x1,1 , x1, 2 ,, x1,36 :
the ages of 36 customers shopping at the inner-city store.
x2,1 , x2, 2 ,, x2, 49 :
the ages of 49 customers shopping at the suburban
store.
Then,
36
x1 
49
x
i 1
1,i
36
 40, x2 
x
i 1
2 ,i
49
 35
and
 x
36
s1 
i 1
The point estimate of
1,i
 x1 
36  1
 x
49
2
 9, s2 
i 1
 x2 
2
2 ,i
49  1
 10 .
1  2  x1  x2  40  35  5 .
1
The point estimate of
 12  s12  92  81 .
The point estimate of
 22  s22  102  100
.
General Case:
1 : the mean of population 1
 2 : the mean of population 2
 12 : the variance of population 1
 22 : the variance of population 2
Let
x1,1 , x1, 2 ,, x1, n1 : the random sample from population 1
x2,1 , x2, 2 ,, x2, n2 : the random sample from population 2.
Then,
n1
x1 
x
1,i
i 1
: the sample mean of population 1.
n1
n2
x2 
x
i 1
n2
2 ,i
: the sample mean of population 2
and
 x
n1
s1 
i 1
1,i
 x1 
n1  1
2
: the standard deviation of population 1.
2
 x
n2
s2 
i 1
 x2 
2
2 ,i
n2  1
: the standard deviation of population 2.
The point estimate of u1  u 2
 x1  x2
Important Properties of X 1  X 2 :
X 1 : the sample statistic with possible value x1
X 2 : the sample statistic with possible value x2
Then,
 X  X  E X1  X 2   1  2
1
2
and

I.
2
X1  X 2
 Var X 1  X 2   E X 1  X 2   1   2  
2
 12
n1

 22
n2
Large Sample Case ( n1  30, n2  30 ):
Sampling Distribution of X 1  X 2 ( n1  30, n2  30 ):

X 1  X 2  N 1   2 ,

X
1
 X 2    X1  X 2
 X X
1
2
X1  X 2

X
1


 12  22 

 N  1   2 ,

n
n
1
2 

 X 2   1   2 



n1 n2
2
1
2
2
2
 N 0,1
Derivation of 1    100% confidence interval:
3
.
1    P Z  z 
2

  X  X      

2
1
2
 P 1
 z 

2
 X1  X 2


       X  X 

2
1
2
 P 1
 z 

2
 X1  X 2





1   2    X 1  X 2 

 P  z 
 z 

2
2
 X1  X 2


 P  z  X1  X 2  1   2    X 1  X 2   z  X1  X 2 
2
2


 P  X 1  X 2   z  X1  X 2  1   2   X 1  X 2   z  X1  X 2 
2
2


Thus, 1     100% confidence interval is
x1  x2   z  X  X  x1  x2   z
2
1
2
2
 12  22

n1 n2

 12  22
 12  22 
  x1  x2  z
 , x1  x2  z
 
2
2
n
n
n
n2 

1
2
1
1    100% confidence interval ( n1  30, n2  30 ):
 As
1,  2
are known,
x1  x2   z  X  X  x1  x2   z
2
1
2
2
 12  22

n1 n2

 12  22
 12  22 
  x1  x2  z
 , x1  x2  z
 
2
2
n
n
n
n2 

1
2
1
is a 1    100% confidence interval estimate of the population
4
1  2 .
difference
 As
1,  2
x1  x2   z
are unknown,
2
s X1  X 2  x1  x2   z
2
s12 s 22

n1 n2

s12 s 22
s12 s 22 
  x1  x2  z
 , x1  x2  z
 
2 n
2 n
n
n2 

1
2
1
is a 1    100% confidence interval estimate of the population
difference
1  2 .
Example (continue):
s X1  X 2
s12 s 22
9 2 10 2




 2.07
n1 n2
36 49
A 95% confidence interval for
x1  x2   z
2
1  2
is
s X1  X 2  40  35  z 0.025  2.07  5  1.96  2.07  5  4.06  0.94,9.06
II. Small Sample Case ( n1  30, n2  30 ):
Two assumptions are made:
1. Both populations have normal distribution.
2. The variance of the populations are equal (  1
2
Pooled estimate of
2
and
 X2
5
1X2
:
  22   2 )
2
 2 , denoted by s p ,
The pooled estimate of
 x
n1
s 2p 
n1  1s
 n2  1s

n1  n2  2
2
1
2
2
i 1
1,i
 x1    x 2,i  x 2 
2
n2
2
i 1
n1  n2  2
is a weighted average of the two sample variance
s12
and
s22 .
The estimate of
 X X
1
2
1 1
 12  22


  2   
n1 n2
 n1 n2 
is
s X1  X 2 
 1
1 
 .
s 2p 

 n1 n2 
1    100% confidence interval ( n1  30, n2  30 ):
x1  x2   t n n 2,
1
2
2
s X1  X 2  x1  x2   t n1  n2 2,
2
1 1
s 2p   
 n1 n2 

1
1 
2 1
2 1




 x1  x2  t n1  n2 2, s p    , x1  x2  t n1  n2 2, s p    
2
2

 n1 n2 
 n1 n2  
is a 1    100% confidence interval estimate of the population
difference
1  2 .
Note:
X 1  X 2  u1  u2  X 1  X 2  1   2 

 T n1  n2  2
s X1  X 2
1 1
s 2p   
 n1 n2 
6
Then, 1    100% confidence interval in small sample case can be
derived analogous to the one in large sample case.
Example:
Let
1 : the mean balance of checking account in Chekry Grove bank
 2 : the mean balance of checking account in Beechmont bank.
x1  1000, n1  12, s1  150
x 2  920, n2  10, s 2  120
Please find a 90% confidence interval for
1  2 .
[solution:]
s
2
p

n1  1s12  n2  1s22

n1  n2  2
11  150 2  9  120 2

 18855 .
12  10  2
Thus,
1 1
1
1
s X1 X 2  s 2p     18855    58.79 .
 12 10 
 n1 n2 
Then, a 90% confidence interval for
x1  x2   t n  n 2,
1
2
2
1  2
is
s X1  X 2  1000  920  t 20,0.05  58.79  80  1.725  58.79
 80  101.41
  21.41, 181.41
Online Exercise:
Exercise 12.1.1
Exercise 12.1.2
7
Related documents