CODE 1 FACULTY OF MATHEMATICS AND INFORMATICS
FINAL EXAM - PROBABILITY AND STATISTICS - Semester 2025.1
Course ID: MI2020E. Time duration: 90 minutes
Problem 1 (2.0 points) An intrusion detection system (IDS) monitors incoming traffic.
On any given hour, the prior probability that there is a real intrusion is 1%. The IDS raises
an alert when it detects suspicious activity. The IDS has the true positive rate (sensitivity)
of 90% and the true negative rate (specificity) of 98%.
a) Find the probability that neither there is a real intrusion nor the IDS raises an alert.
b) Find the probability that the IDS raises an alert.
c) If the IDS raises an alert, what is the probability that there is actually a real intrusion?
Problem 2 (2.0 points) A laboratory is funded to detect a trace element in a food sample.
The laboratory can conduct up to three experiments, with adaptive funding and success
probabilities depending on previous outcomes.
• Experiment 1: Cost: 20 million VND; Probability of success: 0.6.
• Experiment 2 (only if Experiment 1 fails): Additional funding: 25 million VND; Proba-
bility of success: 0.7.
• Experiment 3 (only if Experiments 1 and 2 both fail): Additional funding: 25 million
VND; Probability of success: 0.75.
If the laboratory fails in all three experiments, the project is terminated. Let X be the total
amount of money the laboratory has spent.
a) Find the probability distribution table of X and compute the expected amount of money
spent by the laboratory.
b) Assume that before Experiment 3, the laboratory may choose not to continue, incurring
no further cost but accepting failure. Determine whether continuing to Experiment 3 is
optimal if a successful detection yields a scientific benefit valued at 100 million VND.
Problem 3 (2.0 points) Suppose a company analyzes user engagement during a short ad-
vertising campaign. Let
• X be the normalized engagement score of a user, with 0 < X < 1;
• Y be the normalized cognitive load index, with 0 < Y < 0.5.
Users remain engaged only if the engagement exceeds twice the cognitive load, i.e. X > 2Y .
Assume that the random vector (X, Y ) has the joint probability density function:
c(2x + y) if 0 < 2y < x < 1,
f (x, y) =
0 otherwise.
a) Find the normalizing constant c.
b) Compute the average normalized cognitive load index given that the normalized engage-
2
ment score is .
3
Problem 4 (4.0 points) After an advertising campaign aimed at increasing long-distance
telephone usage, a random sample of 20 household bills was taken. The data as follows:
19.61, 20.14, 19.57, 19.26, 14.03, 19.24, 15.98, 24.85, 26.00, 19.46,
18.29, 16.91, 26.15, 19.64, 16.75, 20.52, 25.47, 18.19, 12.56, 28.47.
Assume that the monthly long-distance telephone bill is normally distributed with a mean of
µ and a standard deviation of 3.87 USD.
a) Find the point estimator of µ by the maximum likelihood estimation method and prove
that it is unbiased.
b) Find a 95% confidence interval estimate of µ.
c) Past experience indicates that before the advertising campaign, the mean monthly long-
distance telephone bill was 17.85 USD. Do the data allow us to infer that the campaign
was successful at level of significance α = 1%?
d) Find the probability of type II error β as a function of µ and α and prove that if the
level of significance α increases then β will decrease.
Given that: critical values of the standard normal distribution N (0, 1): z0.01 = 2.326; z0.05 =
1.645; z0.025 = 1.96 and critical values of Student distribution: t19;0.1 = 1.33; t19;0.05 = 1.73;
t19;0.025 = 2.09.
—————————————- The end —————————————-