1.

Given that a student has an engineering background, what is the probability they score 70 or less in co
mponent 1? (Do not use the data set for this question) [5 Points]
Ans We are given that a student With an engineering background has scores in component one that are
distributed (approximately) as a normal random variable with mean 75 and standard deviation 3 we want t
o find the probability that this core 70 or less in component 1.
We can standardize the distribution of scores using the formula z = (x-) / o, where x is the
score, u is the mean, and o is the standard deviation. In this case, we have:
z = (70-75)/3 = -1.67
We want to find the probability that a student scores 70 or less, which is the same as the
probability that their standardized score is less than or equal to -1.67. We can look up this
probability in a standard normal table or use a calculator to find that:
P(Z < -1.67) = 0.0475
Therefore, the probability that a student with an engineering background scores 70 or less
in component 1 is approximately 0.0475 or about 4.75%.
2. In the given data set, what fraction of students with an engineering background have
scored 70 or less in component 1? [5 Points]
Ans: - We can use the normal distribution to calculate the probability of a student
scoring 70 or less in component 1, given that they have an engineering background.
Let X be the random variable representing the component 1 score of a student with an
engineering background. We know that X follows a normal distribution with mean 75
and standard deviation 3. We can standardize X as follows:
Z=(X-75)/3
The probability that a student scores 70 or less is equivalent to the probability that Z is
less than or equal to (70-75)/3 = -5/3. We can find this probability using a standard
normal distribution table or a calculator:
P(Z < -5/3) ≈ 0.0918
Therefore, about 9.18% of students with an engineering background scored 70 or
less in component 1.

3. Given the distributions, what is the expected value of the class score in component 1?
(Do not use the data set for this question) [5 Points]
Ans:- To calculate the expected value of the class score in component 1, we need to use
the law of total probability. Let E, C, and O represent the events that a student has an
engineering, commerce, or other background, respectively. Then, we have:
P(E) = 0.6, P(C) = 0.3, and P(O) = 0.1
Let X be the component 1 score of a randomly selected student. Then, we can write:
E[X] = E[X | E]P(E) + E[X | C]P(C) + E[X | O]P(O)
where E[X | E], E[X | C], and E[X | O] are the conditional expected values of X given that the
student has an engineering, commerce, or other background, respectively.
From the problem statement, we know that:
- E[X | E]=75
- E[X | C] = 76
- E[X|0] =85
Substituting these values and the probabilities into the equation above, we get:
E[X]=75(0.6)+76(0.3) +85(0.1) = 74.4 +22.8 +8.5=105.7
Therefore, the expected value of the class score in component 1 is 105.7.

4. In the given data set, what is the average score by students in component 1? [5 Points]
Ans: - To find the average score by students in component 1, we need to take into account
the proportion of students from each background and the corresponding mean score and
standard deviation for each background.
Let E, C, and O denote the events that a student is from an engineering, commerce, or other
background, respectively. Then, we have:
P(E) = 0.6
P(C) = 0.3
P(0) = 0.1
Let X denote the component 1 score of a randomly selected student. Then, the overall mean
score is given by:
= E(X) = E(X | E)P(E) + E(X | C)P(C) + E(X | O)P(0)
where E(X | E), E(X | C), and E(X | O) denote the mean scores for students with engineering,
commerce, and other backgrounds, respectively.
Using the given information, we have:
E(X | E) = 75
E(X | C) = 76
E(X | 0) = 85
Substituting these values and the probabilities, we get:
= (75)(0.6) + (76) (0.3) + (85)(0.1) ≈ 75.7
Therefore, the average score by students in component 1 is approximately 75.7.
5. Given that a student scored 80 or more in component 1, what is the probability that this
student is neither from an engineering background nor a commerce background? (Do not use
the data set for this question) [10 Points]
Ans: - Let A be the event that a student is from an engineering background, B be the event
that a student is from a commerce background, and C be the event that a student is from
neither an engineering nor a commerce background.
We are given:
P(A) = 0.6
P(B) = 0.3
P(C) = 0.1
We are also given the following conditional distributions of component 1 scores:
If a student is from an engineering background (A):
X~ N(75, 3^2)
If a student is from a commerce background (B):
Y~ N(76,5^2)
If a student is from neither an engineering nor a commerce background (C):
Z ~ N(85, 4^2)
We want to find P(C | component 1 score ≥ 80). Using Bayes' theorem, we have:
P(C | component 1 score ≥80) = P(component 1 score ≥ 80 | C) *P(C) / P(component 1
score ≥ 80)
We can find each of these probabilities separately:
P(component 1 score ≥ 80 | C) = P(Z ≥80) = 1 - P(Z <80) = 1 -((80-85)/4) ≈ 0.2119
where is the cumulative distribution function of the standard normal distribution.
P[component 1 score ≥80)=P(component 1 score ≥ 80 | A)*P(A) + P(component 1 score
≥ 80 | B) * P(B) + P(component 1 score ≥ 80 | C) *P(C)
Since we don't have the exact value of P(component 1 score ≥ 80 | A) and P(component 1
score ≥ 80 | B), we can use approximations. Since the mean scores for A and B are both less
than 80, it is unlikely that a large proportion of students from A or B will score 80 or more.
On the other hand, the mean score for C is greater than 80, so it is likely that a significant
proportion of students from C will score 80 or more. Therefore, we can approximate:
P(component 1 score ≥ 80 | A) ≈ 0.05
P(component 1 score ≥ 80 | B) ≈ 0.15
Then,
P(component 1 score ≥80) ≈ 0.05* 0.6+ 0.15* 0.3 +0.2119 * 0.1*0.105
Finally, we have:
P(C component 1 score ≥ 80) P[component 1 score ≥ 80 | C) * P(C) / P(component 1
score ≥80) ≈ 0.2119* 0.1/0.105 ≈ 0.2019
Therefore, given that a student scored 80 or more in component 1, the probability that this
student is neither from an engineering background nor a commerce background is
approximately 0.2019.

6. What percentage of the students who have scored over 80 in component 1 are neither from
an engineering background nor a commerce background? [5 Points]
Ans: - To find the percentage of students who have scored over 80 in component 1 and are
neither from an engineering nor a commerce background, we need to use Bayes' theorem
and the law of total probability.
Let E be the event that a student has an engineering background, C be the event that a
student has a commerce background, and O be the event that a student has some other
background. Let S be the event that a student scores over 80 in component 1.
We want to find P(OIS), the probability that a student with some other background has
scored over 80 in component 1.
Using Bayes' theorem, we have:
P(OIS) = P(S|O) *P(0)/P(S)
We can find P(S|O), the probability that a student with some other background scores over
80 in component 1, using the normal distribution with mean 85 and standard deviation 4:
P(S|O) = P(X> 80), where X ~ N(85, 4)
Using a standard normal table or calculator, we can find:
P(Z > (80-85)/4)=P(Z > -1.25) = 0.8944
where Z is a standard normal random variable.
Therefore, P(S|O) = 0.8944.
We can find P(S) using the law of total probability:
P(S) = P(SIE)*P(E) + P(SIC) * P(C) + P([S]O) *P(0)
To find P(SIE), the probability that a student with an engineering background scores over
80 in component 1, we use the normal distribution with mean 75 and standard deviation 3:
P(SIE) = P(X> 80), where X ~ N(75,3)
Using a standard normal table or calculator, we can find:

P(Z> (80-75)/3) = P(Z> 1.67) 0.0475

Therefore, P(S|E) = 0.0475.
To find P(S|C), the probability that a student with a commerce background scores over 80 in component 1
, we use the normal distribution with mean 76 and standard deviation 5:
P(S|C) = P(X >80), where X ~ N(76, 5)
Using a standard normal table or calculator, we can find:
P(Z> (80-76)/5) = P(Z > 0.8) = 0.2119
Therefore, P(S|C) = 0.2119.
We are given that P(E) = 0.6, P(C) = 0.3, and P(0) = 0.1.
Therefore,
P(S)=0.0475*0.6+0.2119 * 0.3 +0.8944 * 0.1 0.19154
Now we can use Bayes’ theorem to find P(OIS):
P(OIS) = P(S|O) * P(O) / P(S)
= 0.8944 * 0.1/0.19154
= 0.467
Therefore, approximately 46.7% of the students who have scored over 80 in component 1 are neither fro
m an engineering background nor a commerce background.

7. The final score obtained by a student is the average of the scores in the three components. Draw a sa
mple of the students by choosing students with serial numbers 1, 11, 21, ... 291. Assume this to be a rand
om sample.
a) Based on this sample, what is a point estimate of the mean score of students taking this course?
b) Based on this sample, what is a 95% confidence interval of the mean scores of students taking this cou
rse? [15 Points (5 + 10)]

8. Suppose that the data set here represents a sample from a population of similar students. Based on th
e data set, would you conclude that students with engineering backgrounds have an average score of 75 i
n component 1? (Use a=0.05.) [15 Points]
Ans: - To determine whether we can conclude that students with engineering backgrounds have an avera
ge score of 75 in component 1, we need to conduct a hypothesis test. Let’s set up the null and alternative
hypotheses:
-Null hypothesis HO: The average score of students with engineering backgrounds in component 1 is 75.
- Alternative hypothesis Ha: The average score of students with engineering backgrounds in component 1
is not 75.
We’ll use a significance level of a=0.05, which means that we’ll reject the null hypothesis if the probability
of observing a sample mean as extreme as the one we obtained (or more extreme) is less than 5%.
Let’s assume that we have a sample of size n, and we compute the sample mean x and the sample stand
ard deviation s. Then, we can calculate the test statistic as:
t = (x-)/(s/sqrt(n))
where u is the hypothesized population mean (in this case, 75). The test statistic follows a t-distribution wi
th n-1 degrees of freedom.
From the data given, we know that the sample size of engineering students is not given, so we cannot co
mpute the test statistic. Hence, we cannot conduct the hypothesis test and cannot conclude whether stud
ents with engineering backgrounds have an average score of 75 in component 1 or not.

9. Suppose we choose two random samples of 30 students each from the class. The first sample contains
students who have commerce backgrounds, and the second sample contains students who have engine
ering backgrounds. The average of the scores of the first sample is 75.8333, with a sample standard devi
ation of 5.7813. The average of the scores of the second sample is 74.7444, with a sample standard devi
ation of 3.4416. Based on these samples, would you conclude that students with commerce backgrounds
scored better than students with engineering backgrounds? (Use =0.05α=0.05.) [15 Points]
Ans: - To determine if students with commerce backgrounds scored better than students with engineering
backgrounds, we need to conduct a hypothesis test. Let u1 and u2 be the true mean scores of commerce
and engineering students, respectively. The null hypothesis is that there is no difference between the mea
n scores of the two groups, while the alternative hypothesis is that there is a difference.
HO: 1 - 2 = 0 (there is no difference between the mean scores of commerce and engineering students)
Ha: 1 - 2 > 0 (commerce students scored higher than engineering students)
We will conduct a two-sample t-test since the population standard deviations are unknown and the sampl
e sizes are less than 30.
We can calculate the pooled variance using the sample standard deviations of the two samples:
s_p^2 = ((n1-1)s1^2 + (n2-1)s2^2)/ (n1+n2-2)
where n1 = n2 = 30
s_p^2 = ((30-1)(5.7813)^2 + (30-1) (3.4416)^2)/ (30+30-2) = 22.2903
The test statistic is calculated as:
t = (x1-x2)/(s_p* sqrt(2/n))
where x1 = 75.8333, x2 = 74.7444, n = 30
t = (75.8333-74.7444) / (sqrt(22.2903) * sqrt(2/30)) = 2.185
The degrees of freedom are (n1 +n2-2) = 58, and the critical value for a one-tailed t-test at α = 0.05 is 1.6
706. Since the calculated t-value (2.185) is greater than the critical value (1.6706), we reject the null hypot
hesis and conclude that there is sufficient evidence to support the claim that students with commerce bac
kgrounds scored better than students with engineering backgrounds in component 1.

10. Based on the average of all three components, grades are awarded to students. The rulebook says th
at no more than 20% of the students can score the highest grade: A, and together, no more than 60% of t
he students can score the highest and second highest grade i.e., A and B. Students who score neither A
nor B get a C grade. Now, the instructor wants to give as many A and B grades as the rule book permits. I
n this context, answer the following questions:
a) What is the average score obtained by students with A grades?
b) What is the average score obtained by students with B grades?
c) What fraction of students with an engineering background have scored A grades?

Ans: - a) To determine the average score obtained by students with A grades, we need to first find the cut
off for an A grade. Since no more than 20% of the students can score an A grade, we can use the normal
distribution to find the score that corresponds to the 20th percentile.
Let X be the random variable representing the component 1 scores for all students. We know that student
s with an engineering background have X ~N(75, 3^2), students with a commerce background have X~ N(
76, 5^2), and students with neither have X ~ N(85, 4^2).
The probability of getting an A grade can be expressed as P(Xza), where a is the cutoff for an A grade. Si
nce no more than 20% of the students can score an A grade, we want to find a such that P(X = a) = 0.2.
Using the normal distribution tables or a calculator, we can find that a = 83.25.
Now we can calculate the average score obtained by students with A grades using the conditional mean f
ormula:
E(XX≥83.25)=([83.25,00] xf(x | X≥ 83.25) dx/P(X = 83.25)
where f(x | X≥ 83.25) is the conditional density function of X given that X ≥ 83.25.
Using the properties of the normal distribution and the fact that the probability density function of X given t
hat X≥ 83.25 is proportional to the probability density function of X truncated at 83.25, we can simplify the
above expression as:
E(XX≥ 83.25)= [[[83.25,∞) x f(x) dx]/[1-((83.25 -)/o)]
where f(x) is the probability density function of X, and σ are the mean and standard deviation of X for stu
dents with neither an engineering nor a commerce background (i.e., X~ N(85, 4^2)), and is the cumulative
distribution function of the standard normal distribution. Plugging in the values, we get:
E(XX≥83.25)= [[83.25,∞) x (0.1/4√(2)) exp[-0.5((x-85)/4)^2] dx]/[1-((83.25-85)/4)] Using a calculator or nu
merical integration software, we can evaluate the integral to get: E(XX83.25) 87.16
Therefore, the average score obtained by students with A grades is approximately 87.16. b) To determine
the average score obtained by students with B grades, we need to find the cutoff for a B grade. Since no
more than 60% of the students can score an A or B grade, we can use the normal distribution to find the s
core that corresponds to the 60th percentile.

Using the same approach as in part a), we can find that the cutoff for a B grade is approximately 81.08.
Now we can calculate the average score obtained by students with B grades using the conditional mean f
ormula:
E(X | 83.25>X≥ 81.08) = [[81.08,83.25] xf(x | 83.25>X≥ 81.08) dx/P(81.08 <X <83.25)
Using the properties of the normal distribution and the fact that the probability density function of X given t
hat 81.08 <X<83.25 is proportional to the probability density function of X truncated at 83.25 minus the pr
obability density function of X truncated at 81.08, we can simplify the above expression as:
E(X 83.25 X 81.08) = [[81.08,83.25] x f(x) dx]/[((83.25 -)/o)-((81.08 -)/σ)]
where f(x) is the probability density function of X, u and σ are the mean and standard deviation of X for stu
dents with neither an engineering nor a commerce background (i.e., X~ N(85, 4^2)), and is the cumulative
distribution function of the standard normal distribution. Plugging in the values, we get:
E(X | 83.25>X≥ 81.08) = [[81.08,83.25] x (0.1/4√(2)) exp[-0.5((x-85)/4)^2 dx]/[((83.25 - 85)/4)-((81.08-85)/4
)]
Using a calculator or numerical integration software, we can evaluate the integral to get: E(X | 83.25>X≥ 8
1.08) = 82.55
Therefore, the average score obtained by students with B grades is approximately 82.55.
c) To find the fraction of students with an engineering background who have scored A grades, we need to
calculate the conditional probability P(A | E), where A denotes the event of scoring an A grade and E den
otes the event of having an engineering background.
Using Bayes' theorem, we can express this probability as:
P(A|E) = P(EA) P(A) / P(E)
where P(EA) is the probability of having an engineering background given that the student scored an A gr
ade, P(A) is the probability of scoring an A grade (which we calculated in part a)), and P(E) is the probabili
ty of having an engineering background (given as 60% in the problem statement).

To find P(E|A), we can use the fact that students with an engineering background have X~ N(75, 3^2), an
d the cutoff for an A grade is 83.25. Thus,
P(EA) = P(X≥ 83.25 | E) = 1 - ((83.25-75)/3)
Using a calculator or the standard normal distribution tables, we can evaluate this probability to get:
P(EA)≈ 0.000232629
Now we can plug in the values to get:
P(A|E)=(0.000232629) (0.2)/(0.6) 0.000077543
Therefore, the fraction of students with an engineering background who have scored A grades is approxi
mately 0.000077543.
d) To find the fraction of students who scored A grades and have an engineering background, we can use
Bayes' theorem again to express this probability as:
P(EA) = P(A| E) P(E)/P(A)
where P(A|E) is the probability of scoring an A grade given that the student has an engineering backgroun
d (which we calculated in part c)), P(E) is the probability of having an engineering background (given as 6
0% in the problem statement), and P(A) is the probability of scoring an A grade (which we calculated in pa
rt a)).
Plugging in the values, we get:
P(EA) = (0.000232629) (0.6)/(0.2)=0.00138777
Therefore, the fraction of students who scored A grades and have an engineering background is approxim
ately 0.00138777, or about 0.14%.