MAT 2377 - Winter 2011

Practice Exam - Solutions

1. X = # of failures B(20, 0.09). We want




20
20
0
20
P (X 2) =
(.09) (.91) +
(.09)1 (.91)19
0
1


20
+
(.09)2 (.91)18
2
=0.73
2. X = # of required mice. X geometric with p = 1/6. We want
P (X = 8) = (1 p)7 p = 0.0465
3. Let A=positive result and M =having the disease. We know that P (A|M ) =
436
5
, P (A|M 0 ) = 500
, P (M ) = 0.077. Using Bayes Formula :
450
P (M |A) =

P (A|M )P (M )
= 0.89
P (A|M )P (M ) + P (A|M 0 )P (M 0 )

4. Let Ai = ith component works. The probability that the circuit is working is

=
=
=
=

P (A1 (A4 (A2 A3 )))

P (A1 )P (A4 (A2 A3 )) ( by independence )
P (A1 )[P (A4 ) + P (A2 A3 ) P (A4 A2 A3 )] ( addition rule )
P (A1 )[P (A4 ) + P (A2 )P (A3 ) P (A4 )P (A2 )P (A3 )] ( by independence )
0.5 [0.5 + (0.4)(0.7) (0.5)(0.4)(0.7)] = 0.32

Thus, the probability that the circuit is not working is 1 0.32 = 0.68.
5.
X = E[X] =

x fX (x) = 1.6

xRX

6.
2

E[X ] =

1
2
x2 (3 x2 ) dx = 0.6 X
= E[X 2 ] 2X = 0.0375

7. This is probability density function for an exponential random variable

with = 1/2. Thus, the mean and the variance of the population are :
= 1/ = 2, 2 = (1/)2 = 4. Hence,
!


81
X


170/81 2

P
Xi > 170
= P X > 170/81 1
4/ 80
i=1
= 1 (0.44) = 0.33
8. X 1 + X 2 N (75 + 80, 288/16 + 162/9) = N (155, 36). Therefore,



156.5 155

P (X 1 + X 2 > 156.5) = 1
36
= 1 (0.25) = 0.4013
9. Since

 2
 2
1
2
1
2
b 1] =
+
2 =
V [
2
2
2

and
b 2 ] = 22 2 + (1)2 2 = 5 2 ,
V [
b 1 ] < V [
b 2 ].
therefore, V [
10. The sample mean and sample standard deviation are
sP
n
2
2
i=1 xi n x
= 0.02558.
x = 1.009 and
n1
Note that t.025,91 = 2.306. A 95% confidence interval for is
s
x 2.306 = [0.990, 1.028].
n
11. X N (31, 0.22 ). We want





31.5 31
31.5 31
P (30.5 < X < 31.5) =

0.2
0.2
= (2.5) (2.5) = 0.9876
2

12. Conditions : normal population with known.

An 95% confidence interval for is
!
2

x z.025 = 98 1.96
= [97.446, 98.554].
n
25
13. We want to test H0 : = 10 against H1 : > 10. We compute
x = 10.03 and s2 = 0.08678.
The observed value of the test statistic is
x 10
10.03 10
=
= 0.322.
t0 =
s/ n
0.08678/ 10
The p-value is p = P (T > .322) which is between 0.25 and 0.40, see
Table V with = n 1 = 9.
14. The observed value of the test statistic is
x 95
94.32 95
=

= 2.27.
z0 =
/ n
1.2/ 16
It is a two-sided test, hence
P = 2P (Z > | 2.27|) = 2 (1 (2.27)) = 0.023. Since P > , thus
we do not reject H0 .

0 . The critical region in terms of its

15. The test statistic is Z0 = X
/ n
x

0 is | x
0 | > z/2 . Thus, the critical region in
observed value z0 = /
n
/ n
terms of x is

x < 0 z/2
or x > 0 + z/2 .
n
n
Therefore the critical region is
p
p
x < 5 1.96(0.5)/ (8) = 4.65 or x > 5 + 1.96(0.5)/ (8) = 5.35

16.
< 5.35 when = 5.1)
(5.1) = P (4.65 < X




5.35 5.1
4.65 5.1

0.5/ 8
0.5/ 8
= (1.41) (2.55) = .9207 .0054 = 0.915.
3

17.
n

hz

i2

.005

(2.576) (30)
=
15

2
= 26.5

Thus, we will collect n = 27 observations.

18. Since the population is normal, then (X )/(S/ n) has a t distribution with = n 1 = 9 degrees of freedom. Hence c = t0.01,9 = 2.821.
19.
Z

1/3

P (Y < 1/3) =
0

6  2 xy 
x +
dy dx = 0.1071
7
2

20.
E[X] =

x f (x, y) = 1; E[X 2 ] =

(x,y)

E[Y ] =

2
x2 f (x, y) = 1.5 X
= 1.512 = 0.5

(x,y)

y f (x, y) = 1; E[Y 2 ] =

(x,y)

E[XY ] =

y 2 f (x, y) = 1.5 Y2 = 1.512 = 0.5

(x,y)

xy f (x, y) = 1.25; XY = 1.25 (1)(1) = 0.25

(x,y)

p
2
Y2 = 0.25/ 0.52 = 0.5
Thus, XY = XY / X
21. Sxx = 59 252 /12 = 6.91667; Sxy = 880.5 (25)2 (432)2 /12 = 19.5;
and Syy = 15648(432)2 /12 = 96. Thus, b1 = Sxy /Sxx = 2.81928 and
b0 = y b1 x = 432/12 (2.81928)(25/12) = 41.8735. The predicted
value response for x = 5 is yb = b0 + b1 (5) = 27.78.
22. X=number of particles in 10kg has a Poisson distribution with mean
= (0.02)(10) = 0.2. We want
P (X = 0) = e0.2

(0.2)0
= 0.819
0!

23. We will use

b2 =

Syy b Sxy
4.1289 (0.83075)(4.4094)
=
= 0.038816
n2
14 2

and

Sxy
4.4094
Sxy
Sxx =
=
= 5.3077
b =
Sxx
0.83075
b

Thus the estimated standard error for b1 is

s
r

b2
0.038816

bb1 =
=
= 0.0855
Sxx
5.3077
24. There is a positive linear tendency. So, it is either 0.49 or 0.98. But the
linear association is weak, so the most reasonable answer is 0.49.
25. Using the CLT,
> 1.5) 1
P (X

1.5 1
p
4/100

!
= 1(2.5) = 10.993790 = 0.00621