0% found this document useful (0 votes)
32 views6 pages

Stock Performance and Food Expenditure Analysis

This document contains the solutions to 6 practice test questions about probability and statistics. The questions cover topics like probability calculations, normal distributions, confidence intervals, and hypothesis testing. Mathematical formulas and calculations are provided as worked solutions for each multi-part question.
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)
32 views6 pages

Stock Performance and Food Expenditure Analysis

This document contains the solutions to 6 practice test questions about probability and statistics. The questions cover topics like probability calculations, normal distributions, confidence intervals, and hypothesis testing. Mathematical formulas and calculations are provided as worked solutions for each multi-part question.
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

TEST 2

Time: 60 minutes

Question 1. (2 marks) Data evidence shows that last year 25% of the stocks in a stock
exchange performed well, 25% poorly, and the remaining 50% performed on average.
Moreover, 40% of those that performed well were rated a “good buy” by a stock analyst, as
were 20% of those that performed on average, and 10% of those that performed poorly.

a. (1 mark) What is the probability that a stock were rated a “good buy"?

b. (1 mark) What is the probability that a stock rated a “good buy" by the stock analyst will
perform well this year?

Solution

Let

E1 : a stock performed well

E2 : a stock performed on average

E3 : a stock performed poorly

A : a stock were rated a “good buy”

Then P  E1   0,25; P  E2   0,5; P  E3   0,25

and P  A | E1   0,4; P  A | E2   0,2; P  A | E3   0,1

a. P  A  ?

P  A  P  E1  P  A | E1   P  E2  P  A | E2   P  E3  P  A | E3 
9
 0, 25.0, 4  0,5.0, 2  0, 25.0,1 
40

b. P  E1 | A  ?

P  E1  A P  E1  P  A | E1  4
P  E1 | A   
P  A P  A 9
Question 2. (2 marks) A recent report in USA Today indicated a typical family of four
spends $490 per month on food. Assume the distribution of food expenditures for a family
of four follows the normal distribution, with a mean of $490 and a standard deviation of
$90.

a. (1 mark) What percent of the families spend more than $30 but less than $490 per month
on food?

b. (1 mark) What percent of the families spend less than $430 per month on food?

Solution

X N   ,  2  with   490,   90

a) P  30  X  490 =?

P  30  X  490 
 490     30   
 F F 
     
 F  0   F  5,11
 0,5  0
 0,5

b) P  X  430  ?

P  X  430 
 430   
 F   F   
  
 F  0,67   0
 1  F  0,67 
 0, 2514
Question 3. (1 mark) The following density function describes the random variable X.

x
 25 ;0 x5

10  x
f  x   ; 5  x  10
 25
0 ; x   0; 10 

Compute the probability that X is less than 7.

Solution

P  X  7
7
  f  x  dx

0 5 7
  f  x  dx   f  x  dx   f  x  dx
 0 5

10  x
0 5 7
x
  0 dx   25 dx  
 0 5
25
dx

 0,82
Question 4. (1 mark) A corporation is considering a new issue of convertible bonds.
Management believes that the offer terms will be found attractive by 20% of all its current
stockholders. Suppose that this belief is correct. A random sample of 130 current
stockholders is taken. What is the probability that the sample proportion is between 0.18 and
0.22?

Solution

P  0, 2
n  130

 
P 0,18  p  0, 22  ?

Since nP 1  P   20,8  5 , we have

 P 1  P  
p N  P; 
 n 
N  0, 2; 0,0352 

Therefore,


P 0,18  p  0, 22 
 0, 22  0, 2   0,18  0, 2 
 F  F 
 0,035   0,035 
 F  0,57   F  0,57 
 F  0,57   1  F  0,57  
 0, 4314
Question 5. (2 marks) A survey of 750 university students found they were paying on
average $108 per week in accommodation costs. Assume the population standard deviation
for weekly accommodation costs is $22.

a. (1 mark) Construct a 90 per cent confidence interval estimate of the population mean.

b. (1 mark) Construct a 95 per cent confidence interval estimate of the population mean.

Solution

n  750
x  108
  22

a. 1    0,9  z /2  z0,05  1,645

90 per cent confidence interval estimate of the population mean is


x  z /2  108  1,3214
n

b. 1    0,95  z /2  z0,025  1,96

95 per cent confidence interval estimate of the population mean is


x  z /2  108  1,5745
n
Question 6. (2 marks) After many years of teaching, a statistics professor computed the
variance of the marks on her final exam and found it to be  2 = 250. She recently made
changes to the way in which the final exam is marked and wondered whether this would
result in a reduction in the variance. A random sample of this year’s final exam marks are
listed here.

57 92 99 73 62 64 75 70 88 60

a. (1 mark) Calculate the sample mean and variance.

b. (1 mark) Can the professor infer at the 10% significance level that the variance has
decreased?

the variance has decreased? =>  2 < 250 => H_1

Solution

a.

x  74
s  14,499  s 2  210,221

b.

  0,1;  02  250; n  10

 H 0 : 2  250

 H1 :  250
2

 n  1 s 2   2
n 1,1   9, 0,9  4,168
2
Reject H0 if
0
2

Now,

 n  1 s 2  10  1 .14, 4992  7,567


 02 250

Hence we fail to reject H0 => reject H1 .

In conclusion, the professor cannot infer that the variance has decreased.

Common questions

Powered by AI

To find P(X < 7), integrate the density function from 0 to 7. The density function is split as: For 0 ≤ x < 5, f(x) = 0, and for 5 ≤ x ≤ 10, f(x) = (10-x)/25. ∫ from 5 to 7 of (10-x)/25 dx = [10x/25 - x²/(2*25)] evaluated from 5 to 7. Substitute and calculate: (70/25 - 49/50) - (50/25 - 25/50) ≈ 0.82. Thus, the probability that X is less than 7 is approximately 0.82 .

By calculating the sample variance of new exam marks (s² = 210.22) against pre-change variance (σ² = 250), you conduct hypothesis testing: H0: σ² ≥ 250 vs. H1: σ² < 250. The test statistic calculated as: χ² = (n-1)s²/σ² = 9*210.22/250 = 7.567 is compared with critical χ² value (4.168 for df=9 and α=0.1). As 7.567 > 4.168, H0 isn't rejected, indicating marking changes don't significantly reduce variance at 10% significance. This test elucidates marking tweaks' statistical insignificance .

Given a sample mean of $108, population standard deviation of $22, and sample size of 750: The standard error (SE) = σ/√n = 22/√750 ≈ 0.803. For a 95% confidence interval, the Z-score is 1.96. Thus, the confidence interval is: CI = x̄ ± Z*(SE) => 108 ± 1.96*0.803 ≈ 108 ± 1.57 Therefore, the 95% confidence interval is approximately [$106.43, $109.57].

The probability that a stock rated a "good buy" performed well is calculated using Bayes' theorem. Given that a stock performed well last year (25% of stocks), performed on average (50%), or performed poorly (25%), and the probabilities of being rated a "good buy" in each category (40%, 20%, 10% respectively), we have: P(A) = P(E1)P(A|E1) + P(E2)P(A|E2) + P(E3)P(A|E3) = 0.25*0.4 + 0.5*0.2 + 0.25*0.1 = 0.09 Thus, using Bayes' theorem: P(E1 | A) = [P(E1)P(A|E1)] / P(A) = (0.25*0.4) / 0.09 = 0.4 / 0.09 = 4/9 ≈ 0.4444 .

First, calculate the probability that a stock is rated a "good buy": P(A) = P(E1)P(A|E1) + P(E2)P(A|E2) + P(E3)P(A|E3) = 0.25*0.4 + 0.5*0.2 + 0.25*0.1 = 0.09. This means there is a 9% likelihood of a stock being rated a "good buy" across any performance category .

The sample mean provides a central tendency measure, whereas the sample variance offers variability insight. Here, decreased variance in marks signifies reduced score variability due to marking changes. By comparing sample variance to historical variance via hypothesis testing, we assess if marking modifications statistically lower score variation. Despite the sample variance (210.22) being lesser than historical variance (250), statistical tests show no significance at 10% level; thus, we can't conclude variance reduction. This illustrates how sample statistics quantify and hypothesize about effects on outlined changes .

For families spending more than $30 but less than $490, calculate probabilities using Z-scores. With mean $490 and standard deviation $90: Z for $30: (30-490)/90 ≈ -5.11 (essentially 0 due to being far in the tail), Z for $490: (490-490)/90 = 0. The cumulative probability for Z=0 is 0.5. The cumulative probability for Z=-5.11 can be approximated by looking up in Z-table (almost 0). Therefore, the probability range is approximately: P($30 < X < $490) = F(0) - F(-5.11) = 0.5 - 0 ≈ 0.5. Thus, 50% of families spend between $30 and $490 per month .

Using a binomial approximation as normal distribution, where p=0.2 and n=130, calculate the standard error: SE = √[p(1-p)/n] = √[0.2*0.8/130] ≈ 0.035. Compute Z-scores for p̂=0.18 and p̂=0.22: Z(p̂1) = (0.18-0.2)/0.035 ≈ -0.57, Z(p̂2) = (0.22-0.2)/0.035 ≈ 0.57. Using a standard normal table or calculator for Z-scores, obtain P(-0.57 < Z < 0.57) ≈ 0.4314. Thus, the probability the sample proportion is between 0.18 and 0.22 is 0.4314 .

Larger sample sizes generally decrease the margin of error, tightening confidence intervals. In this survey, n=750 yields SE = σ/√n = 22/√750 ≈ 0.803. Larger n decreases SE, thus narrowing the confidence interval for the mean $108, reducing estimation uncertainty. For 95% confidence, CI calculated as 108±1.96*0.803 yields narrower bounds: [106.43, 109.57]. Thus, expansive sample sizes advantageously provide precision and reliability in estimating population parameters, affirming CI's robustness with decreased parameter variance .

The null hypothesis states the variance has not decreased (H0: σ² ≥ 250), while the alternative hypothesis claims it has decreased (H1: σ² < 250). Observed sample variance s² is 210.22 with a χ² statistic calculated as (n-1)*s²/σ₀² = 9*210.22/250 = 7.567. Using χ² distribution tables or calculators, obtain the critical value χ²₀.1,9 = 4.168 for df=9 and α=0.1. Since 7.567 > 4.168, we fail to reject H0. Thus, there is no statistical evidence at the 10% significance level that the variance has decreased .

You might also like