Name:

AMAT 370: Probability for Scientists and Engineers

Midterm
November, 2024

Please answer the questions in the space provided. You have the entire class period to finish the exam. If
you can express your response entirely in terms of the corresponding cumulative distribution function
then you do not need to simplify the appropriate substitutions into a tabular value unless otherwise noted.

Question Points Bonus Points Score

1 0 0
2 0 0
3 0 0
4 0 0
5 0 0
6 0 0
7 0 0
8 0 0
9 0 0
10 0 0
11 0 0
12 0 0
Total: 0 0
1. The variation in a certain electrical current source X (in milliamps) can be modeled by the probability
distribution function
5 1
f (x) = { − x for 2 ≤ x ≤ 4 }
4 4
1. Compute the probability P (3 ≤ X ≤ 3.5) or the probability that the variation in a certain electrical
source is between 3 and 3.5 milliamps.

2. What is the 55th percentile for variation in a certain electrical current source?

3. Compute the average current from this source, E(X).

4. If this current passes through a 220-Ω resistor, the resulting power (in microwatts) is given by the
expression h(X) = (current)2 (resistance), where the resistance in this notation is 220. Compute the
average current from this source as it passes through a 220-Ω resistor.
HINT: express h(X) as a function of X and use the law of the unconscious statistician.
2. Let X = IQ of a randomly selected individual. Assume that X is normally distributed with µ = 100 and
σ = 15. The Mensa organization only allows the top 2.28 % of the population into its society for people with
high IQs. Answer the following questions.

1. By showing your work, find the value of X such that, if X is greater than or equal to this value, then
one is admitted into the Mensa organization.

2. Compute the 84th percentile of IQ scores (You will need to use the standard normal table for this
question).

3. Given your work in item 2. above, compute the IQ score that determines the 84th percentile.
 13
3. 1. Completely simplify Γ( )
2
2. Use the formula
n
Y (2n)!
(2i − 1) =
2n n!
i=1

and the work in part 1. to compute the precise value of the integral
Z ∞ 6
x
√ e−x dx
0 x
HINT: Look at the integrand carefully.
n
Y
1. Remark: The notation f (i) = f (1)f (2) · · · f (n), that is, the product of the formula f (i) evaluated for
i=1
i from 1 through n, so the formula in the problem computes the product of the first n odd numbers.)
2. Remark: You can leave your answers in terms of factorials and products.
4. The frequency distribution of navigation positive errors is approximated by an exponential distribution. Let
X = the lateral position error in nautical miles, which can be either negative or positive. Suppose the
probability distribution function of X is

f (x) = .1e−.2|x|

for −∞ < x < ∞

1. Z
Verify this function is a probability density function, by showing your work, by demonstrating that
∞
f (x)dx = 1
−∞

(HINT: remember that |x| is a piece-wise function, so that will affect computing the integral over the
specified range. Make sure to apply its definition properly and the answer reduces to just two easy calc
II improper integral problems.)

2. Compute P (−2 ≤ X ≤ 2)
5. Products that are subject to wear and tear are generally tested for endurance in order to estimate their useful
lifetimes, in particular, ball bearing products. The lognormal distribution is a model for times until failure
of a ball bearing. Suppose that the engineers are interested in knowing how long to wait until there is a 90
percent chance that a ball bearing will have failed. Let X = the time until a failure of a ball bearing and
suppose its lognormal distribution has parameters µ = 4.15 and σ 2 = .53342 .

1. Compute the 90th percentile of the time to failure of a ball bearing (You will need the standard normal
table for this question).

2. Compute the expected value of the time until a failure of ball bearing.

3. Compute the variance in this experiment.

6. A common allegation is that more babies are born during certain phases of the moon. The following data
gives the proportion of births under the given lunar cycle:

p1 =0.0343 during the new moon

p2 =0.0456 during the f irst quarter moon
p3 =0.2146 during the f ull moon
p4 =0.2132 during the third quarter moon

Define the expected count during the ith phase of the moon to be,

ECi = 30, 703 · pi

for i = 1, 2, 3, 4.
Last, suppose that the following number of births were observed during each moon phase:

7680 during the new moon

7579 during the f irst quarter moon
7711 during the f ull moon
7733 during the third quarter moon

Denote the number of births observed during the ith phase of the moon by OCi , then the statistic

4
X (OCi − ECi )2
X2 =
ECi
i=1

has a χ2 -distribution with 3 degrees of freedom. Answer the following questions.

1. Compute X 2 by expanding the sum appropriately. You may then denote the sum appropriately to use
below.

2. Compute the probability that X 2 is less than or equal to the value you computed in item 1. above. You
do not need to simplify this answer.
7. Let X denote the flying time in hours of an airplane passenger flying from the United States to Europe.
Suppose that X has a Γ distribution with mean value 13 hours and standard deviation of 3 hour.

1. Compute the parameters α and β for X as above.

2. What is the probability that the travel time of a passenger exceeds 15 hours?

3. What is the probability that 14 ≤ X ≤ 16? You do not need to simplify this answer.
8. Define the function hX (x) to be the hazard function of the continuous random variable X. Recall our notation
for the probability distribution function and cumulative distribution function of X is f (x) and F (x), respec-
tively. Given our recollection, the hazard function may be expressed in terms of both f and F in the formula

f (x)
hX (x) =
1 − F (x)
d
1. Verify by computation or citing the relevant rule in calculus that hX (x) = − log(1 − F (x))
dx
2. Suppose that X ∼ W eibull(α, β), compute hX (x).

3. X ∼ W eibull(2, 1), compute hX (100).

9. A production plant produces items that have a probability x of being defective. The plant manager does
not know x, but from past experience expects its value to be 33 % or 1 in 3. Furthermore, she quantifies
her uncertainty about x by attaching a variance of 5.5% or 1 in 18 to her 33% estimate. After consulting
with an expert, the manager decides to use a beta distribution to model x by defining the continuous random
variable X = the probability that the produced item is defective. Given the description of X, the manager
asserts the probability distribution function of X is necessarily the standard beta distribution, that is, the
beta distribution with probability distribution parameters A = 0, B = 1.

1. Given that X ∼ Beta(α, β, 0, 1), use the description of the manager’s work to compute the missing
parameters α and β.
HINT: Use the hypotheses in the question to solve for the missing parameters using the formulas for the
expected value and variance of the beta distribution. This is not entirely easy, so I encourage you to
practice this at home before the test.
2. Compute the probability that X is between 7 % and 10 %.

HINT: I set this problem up so the missing parameters will make this step easy.
10. Consider a pair of discrete random variables (X, Y ) together with their joint probability mass function

X\Y 0 1 2 3
0 1/8 1/8 0 0
1 0 2/8 2/8 0
2 0 0 1/8 1/8

where the notation X\Y means the leftmost column corresponds to the values of X and the top row corresponds
to the values of Y . Further, the (i, j) entry in the table gives the probability p(x = i, y = j), for 0 ≤ i ≤ 2, 0 ≤
j≤3

1. Compute the marginal probability mass functions for both X and Y .

2. Compute the probability P (X ≤ 1).

3. Compute the probability P (Y > 1).

4. Compute the probability P (X + Y ≤ 3).

5. Compute the conditional probability mass function of Y given that X = 2.

11. Let the joint probability distribution function of the continuous random variables (X, Y ) be f (x, y) = 1, where
0 ≤ x ≤ 1 and 0 ≤ y ≤ 1, and 0 otherwise.

1. Compute the marginal pdfs of both X and Y , that is, compute fX (x) and fY (y).

2. Compute µX , σX , µY , σY ,that is, the means and standard deviations of both X and Y .

3. Compute E(XY ).

4. Using the Proposition that Cov(X, Y ) = E(XY ) − µX µY compute the covariance of (X, Y ).

Cov(X, Y )
5. Using the definition that ρX,Y = Corr(X, Y ) = , compute the correlation of (X, Y ).
σX σY
12. Consider the experiment of tossing a fair coin 900 times. For i = 1, . . . , 900 define the sequence of random
1 1
variables Xi = 1 if a H is obtained, Xi = 0 if a T is obtained. By definition, E(Xi ) = and V (Xi ) = .
2 4
1
Accordingly, the {Xi } form a random sample of size n = 900 in an experiment with mean µ = and variance
2
900
1 X
σ 2 = . What is the probability that the total number of heads, defined to be the random variable To = Xi
4
i=1
is greater than 495? That is, compute P (To ≥ 495).

(HINT: Proceed by the Theorem I make a big deal out of in lecture.)