0% found this document useful (0 votes)
58 views9 pages

Chapter 11

This chapter discusses confidence intervals for the parameter of an exponential distribution and the parameters of two exponential distributions. It also covers properties of confidence intervals, such as how their length depends on the confidence level. Additionally, it examines test statistics for comparing the means of two normal populations and the binomial proportion. Key results include formulas for the confidence interval of a single exponential parameter and the difference between two exponential means. The chapter also shows that the length of confidence intervals increases as the confidence level decreases.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
58 views9 pages

Chapter 11

This chapter discusses confidence intervals for the parameter of an exponential distribution and the parameters of two exponential distributions. It also covers properties of confidence intervals, such as how their length depends on the confidence level. Additionally, it examines test statistics for comparing the means of two normal populations and the binomial proportion. Key results include formulas for the confidence interval of a single exponential parameter and the difference between two exponential means. The chapter also shows that the length of confidence intervals increases as the confidence level decreases.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Chapter 11

11.1 P (0  θ  kx )  1  α
 θ
 px  
 k


1 
 θe
 x /θ
dx   e  x /θ  e1/ k  1  α
θ /k
θ/k
1 1
  ln(1  α ) and k 
k ln(1  α )

11.2 (a) p[0  θ  k ( x1  x2 )]  1  α


 θ
p  ( x1  x2 )    1  α
 k
 θ
p  ( x1  x2 )    α
 k
1 θ2 1 1 1 1
  α α k2  k
2 k2 θ2 2k 2 2α 2α

 θ
(b) p  x1  x2    1  α
 k
2
1 θ  1
  2θ  2  1  α
2 k θ
2
1  1
  2   2(1  α ),  2   2(1  α )
k k
1
k
2  2(1  α )
1
k
2  2(1  α )

162
Chapter 11 163

11.3 p( R  θ  cR )  1  α
θ 
p   R  θ   1 α
c 
θ θ
2 2  R2 
θ 2 θ/ c
(θ  R )dR  1  α θ R  
θ2  2 
θ /c
2  2 θ2 θ2 θ2 
θ    2   1 α
θ 2  2 c 2c 
2 1
1  2  1  α , c 2  2c  1  (1  α )c 2
c 2c
2  4  4α 1  1  α
ac 2  2c  1  0 and c  
2α α

11.4 By inspection
zα /3  zα /2  zα /2  z2α /3
2 zα /2  zα /3  z2α /3

length of first confidence interval is less than that of 2nd confidence interval

11.5 Length of confidence interval:


σ  σ 
L  X  z(1 k )α    X  zk α  
n  n
σ
 ( z(1 k )α  zkα ) 
n
σ
If k  1 / 2, L1/2  2 zα /2 
n
If k  1 / 2,
zkα  za /2  δ1  zα /2 δ1  0; z(1 k )α  z(1 k )α  δ 2  z α /2 where δ 2  0

σ
and Lk  [2 zα /2  (δˆ1  δˆ2 )] 
n

Since the normal density function f ( x ) is decreasing for x  0, δ 2  δ1 , thus


σ
Lk  2 zα /2 
n

σ
By the symmetry of f ( x ) , for k  1 / 2 , Lk  2 zα /2 
n
164 Mathematical Statistics, 8E

 σ 
11.6 p  x  μ  zα /2   1 α 
 n 
σ σ
zα /2   E and n  zα /2 
n E
2
 σ
n   zα /2  
 E

s σ
11.7 Substitute tα /2,n 1  for  zα /2 
n n
If x , the mean of a random sample of size n from a formal population with the mean μ , is
used as an estimate of μ , we can assert with (1  α )100% confidence that the error is less than
s
tα /2,n 1  .
n

11.8 If x1 and x2 are the means of independent random samples of size n1 and n2 from normal
populations with μ1 , μ 2 , σ1 , and σ 2 , and x1  x2 is to be used as an estimate if μ1  μ2 , the
probability is 1  α that error will be less than

σ12 σ 22
zα /2 
n1 n2

n1  1 n2  1 n n 2 2
11.9 E ( S 2p )  σ2  σ 2  1 2 σ  σ 2
n1  n2  2 n1  n2  2 n1  n2  2
therefore unbiased

(n1  1) s12 (n2  1) s22


 χ 2 ( n1  1)  χ 2 ( n2  1)
σ 2
σ 2

(n1  1) s12 ( n2  1) s22


  χ 2 ( n1  n2  2) var is 2(n1  n2  2)
σ 2
σ
2

( n1  1) s12  (n2  1) s22 var is 2σ 4 ( n1  n2  2)

(n1  1) s12  ( n2  1) s22 2σ 4


var is
n1  n2  2 ( n1  n2  2)

Z ( x  x )  ( μ 2  μ1 ) 1
11.10 T   1 2 
Y
σ
1 1
 ( n1  n2  2) S 2p
n1  n2  2 n1 n2 n1  n2  2
( x1  x2 )  ( μ1  μ 2 )

1 1
Sp 
n1 n2
Chapter 11 165

11.11  zα /2 nθ (1  θ )  x  np and zα /2 nθ (1  θ )  x  np
z 2α /2 nθ (1  θ )  ( x  nθ )2  x 2  2 xnθ  n 2θ 2
n 2θ 2  nzα2 /2  2 xnθ  nzα2 /2θ  x 2  0
x2
( n  zα2 /2 )θ 2  (2 x  zα2 /2 )θ  0
n
by quadratic formula
 x2 
2 x  za2/2  (2 x  zα2 /2 ) 2  4(n  zα2 /2 )  
 n 
θ

2( n  zα2 /2 
x  nθ  x  nθ 
11.13  zα /2  ;  zα /2
nθ (1  θ  ) nθ (1  θ )
1
Let θ * = value of θ with θ   θ  θ  closest to . By Theorem 11.7,
2
θ * (1  θ *) z2
e  zα /2 and n  θ * (1  θ *) α2/2
n e

11.15 By Theorem 11.8 with probability approximately 1  α


θˆ (1  θˆ1 ) θˆ2 (1  θˆ2 )
E  zα /2 1 
n1 n2

θˆ1 (1  θˆ1 )  θˆ2 (1  θˆ2 )


11.16 If n1  n2  n , then E  zα /2
n
1
The right-hand side of this inequality is maximized when θ1  θ 2  .
2
2 2
1 z z
Thus, E  zα /2 , E 2  α /2 , and n  α /22 .
2n 2n 2E

1 2 1 2
11.17 χα ,2( x 1)  χ 0.01,8  0.050
2n 400

1 σ12 s22 1
11.18  
f1α /2,n1 1,n2 1 σ 2 s1
2 2
fα /2,n1 1,n2 1
s12 1 σ12 s12 1
   2
s2 fα /2,n1 1,n2 1 σ 2 s2 f1α /2,n1 1,n2 1
2 2

s12 1 σ12 s12


    fα /2,n2 1,n1 1
s22 fα /2,n1 1,n2 1 σ 22 s22
166 Mathematical Statistics, 8E

σ σ
11.19 σ  zα /2  s  σ  zα /2
2n 2n
 zα /2   zα /2 
σ 1    s  σ 1  
 2n 2n 
1 1 1
 
 z  s  z 
σ 1  α /2  σ 1  α /2 
 2n   2n 
s s
σ 
z z
1  α /2 1  α /2
2n 2n

9.4 1.96(9.4)
11.20 n = 150 σ  9.4 E  1.96   1.50
150 12.247

9.4
11.21 61.8  2.575   61.8  1.98, 59.82  μ  63.78
150

10.5 10.5
11.22 E  2.575   2.575   2.47 mm
120 10.955

10.5 10.5
11.23 141.8  2.33   141.8  2.33  141.8  2.23
120 10.955
139.57  μ  144.03

s 45
11.24 x  z0.005 ; 52.80  2.575 , or (51.35, 54.25).
n 64

s 2.68
11.25 e  z0.025  1.96  0.83 min.
n 40

σ N n 9.4 900  150


11.26 e  z0.025  1.96  1.37.
n N 1 150 900  1

σ N n 9.4 900  150


11.27 x  z0.005 ; 61.8  2.575 , or (60.01, 63.61).
n N 1 150 900  1

2 2
 σ  12.2 
11.28 n   z0.025   1.96  91.48 or 92, rounded up to the nearest integer.
 e  2.5 

 σ
2 2
 3.2 
11.29 n   zα /2   1.96   354.04 or 355, rounded up to the nearest integer.
 e  1 / 3 
Chapter 11 167

s 0.29
11.30 x  t0.025,n 1 ; 5.68  2.262 , or (5.47, 5.89)
n 10

s 2.75
11.31 x  t0.005,17 ; 63.84  2.898 ; or (61.96, 65.72).
n 18

s 0.625
11.32 e  t0.025,11  2.201  0.40
n 12

σ12 σ 22 4.82 3.52


11.33 ( x1  x2 )  z0.05  ; 5.2  1.645  , or ( 7.49, 2.91).
n1 n2 16 25

s12 s22 19.4 2  18.82


11.34 ( x1  x2 )  z0.05  ; 7.4  2.575 , or ( 16.31,1.51).
n1 n2 61

11(1.2)2  14(1.5) 2
11.35 s 2p   1.8936 s p  1.376
25
1 1
(13.8  12.9)  2.060(1.376) 
12 15
0.9  2.8346(0.387), 0.9  1.098
0.198  μ1  μ 2  1.998 feet

11.36 x1  8260, s1  251.89, x2  7930, s 2  206.52


4(251.89)2  4(206.52) 2
s 2p   53,049.54 s p  230.32
8
1 1
8260  7930  3.355(230.32) 
5 5
330  488.75
158.75  μ1  μ 2  818.75 million calorie per ton

(0.004)2 (0.005)2
11.37 E  2.33 
35 45
 2.33(0.001)  0.0023 ohm

204
11.38 θˆ   0.68
300
(0.68)(0.32) 0.68  0.053
0.68  1.96
300 0.627  θ  0.733

(0.68)(0.32)
11.39 e  2.575  0.069
300
168 Mathematical Statistics, 8E

190 (0.76)(0.24)
11.40 (a)  0.76 0.76  2.575
250 250
0.76  0.070 0.690  θ  0.830

1 190(60) 1
190  (2.575)2  2.575  (2.575)2
(b) 2 250 4
250  (2.575) 2

190  3.315  2.575 45.6  1.658


250  6.631
193.315  17.702
0.684  θ  0.822
256.631

(0.76)(0.24)
11.41 e  1.96  0.053
250

(0.18)(0.82)
11.42 0.18  2.575 0.18  0.099
100
0.081  θ  0.279

54 (0.45)(0.55)
11.43  0.45 e  1.645  0.075
120 120

(0.34)(0.66)
11.44 0.05  z 0.05  0.02735z z  1.83
300
confidence is 2(0.4664)  100  93.3%

(1.96)2
11.45 n   2401
4(0.02)2

2
 1.96 
11.46 n  (0.03)(0.70)   (0.21)(9604)  2017
 0.02 

(2.575)2
11.47 n   1037 rounded up
4(0.04)2

2
 2.575 
11.48 n  (0.65)(0.35)   943
 0.04 
Chapter 11 169

84 156
11.49  0.336  0.624
250 250
(0.336)(0.664) (0.624)(0.376)
(0.336  0.624)  1.96 
250 250
0.288  0.084 0.372  θ1  θ 2  0.204

48 68
11.50  0.096,  0.170
500 400
(0.096)(0.904) (0.170)(0.830)
0.096  0.170  2.575 
500 400
0.074  0.059 0.133  θ1  θ 2  0.015

(0.096)(0.904) (0.170)(0.830)
11.51 e  2.33 
500 400
 2.33(0.022939)  0.053

(1.96)2
11.52 n   769
2(0.05)2

9(0.29)2 9(0.29)2
11.53  σ2 
19.023 2.700
0.04  σ  0.28
2

11(0.625)2 11(0.625)2
11.54  σ2 
19.675 4.575
0.2184  σ  0.939
2
0.47  σ  0.97

4.5 4.5
11.55 σ  3.67  σ  5.83
2.575 2.575
1 1
128 128

2.68 2.68
11.56 σ  2.13  σ  3.62
2.33 2.33
1 1
80 80

19.4 2 1 σ12 19.4 2


11.57     f 0.01,60,60
18.82 f 0.01,60,60 σ 22 18.82
19.4 2 1 σ12 19.42 σ12
    1.84 0.58   1.96
18.82 1.84 σ 22 18.82 σ 22
170 Mathematical Statistics, 8E

(1.2)2 1 σ 2 (1.2)2
11.58   12   f 0.01,14,11
(1.5) f 0.01,11,14 σ 2 (1.5)2
2

0.64 σ 12 σ12
  (0.64)(4.30) 0.165   2.752
3.87 σ 22 σ 22

(251.89)2 1 σ12 (251.89)2


11.59     f 0.05,4,4
(206.52)2 f 0.05,4,4 σ 22 (206.52)2
1.4876 σ 12 σ12
 2  1.4876(6.39) 0.233   9.506
6.39 σ2 σ 22

11.60 Using MINITAB we enter the data into C1 and we give the command
MTB> Tinterval 95.0 C1
Obtaining
N MEAN STDEV SEMEAN 95.0 PERCENT C.I.
20 6.145 1.467 0.328 (5.458, 6.832)

11.61 Using MINITAB we enter the data into C1 and C2 and we give the command
MTB> St Dev C1 obtaining
ST DEV = 275.87

Then, with χ 0.05,29


2
 42.557 and χ 0.95,29
2
 17.70 , we have

29(275.87)2 29(275.87) 2
 σ2 
42.557 17.78

or 227.7  σ  352.3 with 90% confidence.

You might also like