Вы находитесь на странице: 1из 3

2010 AJC H2 MA Statistics P2 Prelim soln

6 No of ways to invite her guests = (2)7 – 1 = 127

No of ways to arrange a round table = 2! X 4! X 2! X 8 = 768

No of ways to select 2 facilitators and 2 teams (Without any restrictions)


= 8C2 x (6C3 x 3C3)  2! = 280
No of ways to select the 2 facilitators and 2 teams with Lee family forming 1 team
= 4C3 x 5C2 x 3C3 = 40

No. of possible formations = 280 – 40 = 240

7 (a)(i) For x on y, x  0.3085 y  31.13  x  0.309 y  31.1 (3 s.f.)


For y on x, y  2.8526 x  95.999  y  2.85 x  96.0 (3s.f.)

(ii) Since chemical Y is the controlled variable, use regression line of x on y.


0  0.3085 y  31.13  y  100.91
The estimation is not valid as this is an extrapolation, linear relation may not hold outside the
range of data.

1(b)(i) By comparing the linear product moment correlation for the 3 models, Model C is the
most appropriate with the highest value of r  0.993 as it best describes the data given.

Using linear transformation w  ln x ,


Regression line of w on y is w  0.026136 y  3.8294 � w  0.0261 y  3.83 (3 s.f.)

(ii) Change in w = 0.026136(5)  0.13068 �0.131 (3 s.f.)


w decreases by 0.131.

(b)(iii) r 2    0.9 2  bd where b  1.5 Hence d 


  0.9 2  0.54
 1.5
Since  
x , z lies on the reg line z  1.5 x  12 ,
z  1.5 x  12
29.1  16.2  8.9  5.1  3.8
x  3.3220655
5
z  1.5(3.3220655)  12  7.0169
Hence, reg line of x on z is x  (3.3220655)  0.54( z  7.0169)
\ x  0.54 z  7.11 (3.s.f)

8 80
(i) Unbiased estimate of m is x   12  12.4
200
200 �
1425 80 2 �
Unbiased estimate of s 2 is s2  ( ) � 7
199 �
�200 200 �
(ii) H0 : m = 12
X  12
H1 : m �12 Test Statistic : Z  ~ N(0,1) under H0 by CLT
s / 200
Using GC, p -value = 0.0325 Given that H0 is not rejected, a < 3.25
(iii) Since we will be using a one-tailed test in stead of a two-tailed test,
1
a < (3.250944)  1.625472 � a < 1.63
2
(iv) The management must have a sampling frame (the list of all cars parked). If there are N cars,
Page 1 of 3
choose a random number, k, from 1 to N/200 (take the nearest integer value), then select every
(N/200)th car until a sample of 200 is obtained.

9 1
P(C) = p �1   1  p  �
(i)
5
(ii) K’C represents the event where Alice does not know the correct answer but she
answers correctly.
1
P ( K '�C ) 1  1 p � 1
 5 
(iii) P(K’|C) = solving, get p =0.75.
P (C ) 16 1
p �1   1  p  � 16
5
When p = 0.3, P(C) = 0.3+0.70.2=0.44

(0.44)3 (0.56)2 �3 3
P(3 consecutive correct | 3 correct answers) = 3 2 5

(0.44) (0.56) �C3 10

P(negative score) = P(0 correct or 1 correct) = (0.56)5  C15 (0.56) 4 (0.44) = 0.271

10 Let X be the r.v. denoting the number of orders for herbal chicken soup in 30 min.
X ~ Poi (2.3)
Let Y be the r.v. denoting the number of orders for herbal chicken soup in 1 hour.
Y ~ Poi (2.3 �2) i.e. Y ~ Poi (4.6)

(i) P  6 �Y < 10   P  Y �9   P  Y �5   0.29471 �0.295(3.s. f )


(ii) Y ~ Poi (4.6) , Mean = variance = 4.6
4.6
Since n = 100 is large, by CLT, Y : N (4.6, ) approximately
100
P(Y > 5)  0.031090 �0.0311(3.s. f )

(iii) Let T be the r.v. denoting the number of orders for herbal chicken soup in a day (8 hours).
T ~ Poi (2.3 �16) i.e. T ~ Poi (36.8)
Since l > 10 , T ~ N (36.8,36.8) approximately.
P  20T �1.4  8T  250  
� 350 �
 P  8.8T  350 �0   P �
T � � P(T �39.773)
� 8.8 �
= P(T40) = P(T>39.5) cc
= 0.328

(iv) Let A be the r.v. denoting the number of orders for herbal chicken soup in 2 hours
(lunch time)
A ~ Poi (2.3 �4) i.e. A ~ Poi (9.2)
Let H be the r.v. denoting the number of days with more than 10 orders during lunch period
from 12 to 2pm.
P  A > 10   1  P  A �10   1  0.68202  0.31798
H ~ Bin  25, P  A > 10  
i.e. H ~ Bin  25, 0.31798 
P ( H < 14)  P( H �13)  0.98947  0.989 (3.s.f)
Since the probability of having less than 14 days with more than 10 orders during the lunch
period is quite high, he should close down his business.
(iv) The use of the Poisson model is not suitable in this context as the number of orders during
Page 2 of 3
lunch and dinner period will likely be higher than the rest of the hours in a day, thus it is
unlikely that the mean number of orders is the same for each 30 minute period.
11 (i) Let A be the r.v. denoting the amount of time taken by the machine to produce a large tub of
ice- cream. A : N ( m , s 2 )
P ( A < 60)  0.88 P ( A > 50)  0.70
� 60  m � � 50  m �
P �Z <  0.88 P �Z >  0.70
� s � � � s � �
60  m 50  m
 1.17499  0.52440
s s
m  60  1.17499s    (1) m  50  0.52440s    (2)

Equating (1) and (2), \s  5.8845  5.88 (3.s.f) and m  53.08583  53.1 (3.s.f)

(ii) Let B be the r.v. denoting the amount of time taken by the machine to produce a small tub of
ice- cream. B : N (20, 22 )

Required prob = P A1  A2  A3  A4  A5  3  B1  B2  B3  > 60 
Let T  A1  A2  A3  A4  A5  3  B1  B2  B3  .
T : N  85.42915, 281.1367 

 
P A1  A2  A3  A4  A5  3  B1  B2  B3  > 60  P  T > 60 

 1  P  60 < T < 60   0.935317  0.935 (3.s.f)

(iv) Let C be the number of times the machine takes at least 60 minutes to produce a large
tub of ice-cream.
C : Bin  n, P  A �60   i.e. C : Bin  n,1  0.88 
Since n is large, np > 5 and n(1-p) > 5, C : N  0.12n, 0.1056n 
P  C > 20  < 0.2
P  C > 20  0.5  < 0.2 (cc)
� 20.5  0.12n �
P �Z > �< 0.2
� 0.1056n �
Using GC, P  Z > 0.84162   0.2
20.5  0.12n
Thus, > 0.84162 � 0.12n  0.2735 n  20.5 < 0
0.1056n
Solving -14.26 < n < 11.98 greatest n = 143.

Page 3 of 3

Вам также может понравиться