Probability and Statistics: Key Concepts, Applications, and Problem Solutions

Probability and Statistics: Core Concepts and Applications

Exam Structure and Instructions

This document provides solutions and explanations for a college-level exam in probability and statistics. The exam covers fundamental concepts such as probability distributions, descriptive statistics, and applications to real-world scenarios (e.g., reliability, insurance, and quality control).

• Duration: 170 minutes, closed book, calculators allowed.

• Format: Direct answers on the questionnaire; clear identification required.

• Scoring: 40 points total; questions require both calculation and explanation.

Probability Distributions and Reliability

Binomial Distribution and System Reliability

Reliability analysis often uses the binomial distribution to model the probability that a certain number of components (e.g., engines) function properly.

• Binomial Distribution: The probability of exactly k successes in n independent Bernoulli trials with success probability p is given by:

$P(X = k) = \binom{n}{k} p^k (1-p)^{n-k}$

• Example: For an aircraft with 2 engines, each with a failure probability of 1/3, the probability that at least 1 engine works (i.e., the flight is safe) is:

$P(X \geq 1) = 1 - P(X = 0) = 1 - (1/3)^2 = 8/9 \approx 0.8889$

• For a 4-engine aircraft, the probability that at least 2 engines work is:

$P(Y \geq 2) = P(Y=2) + P(Y=3) + P(Y=4)$ $= C_4^2 (2/3)^2 (1/3)^2 + C_4^3 (2/3)^3 (1/3) + (2/3)^4$ $= 0.8889$

• Conclusion: Both configurations have the same probability of safe completion under these assumptions.

Descriptive Statistics

Measures of Central Tendency and Dispersion

Descriptive statistics summarize data using measures such as mode, median, percentiles, and the coefficient of variation.

• Mode: The value(s) with the highest frequency in a data set.

• Median: The middle value when data are ordered. For even-sized samples, it is the average of the two central values.

• Percentile: The value below which a given percentage of observations fall. The pth percentile is found by:

$h = p \cdot n / 100$ If h is not an integer, interpolate between the closest ranks.

• Coefficient of Variation (CV): A standardized measure of dispersion, calculated as:

$CV = \frac{s}{\bar{x}} \times 100$

• Where $s$ is the standard deviation and $\bar{x}$ is the mean.

Example Table: Frequency Distribution

MODALITÉ	FRÉQUENCE ABSOLUE
0	2
1	1
2	2
6	1

Interpretation: The modes are 0 and 2, as they have the highest frequencies.

Probability in Insurance and Classification

Conditional Probability and Law of Total Probability

Insurance problems often require calculating the probability of an event given class membership, using the law of total probability and Bayes' theorem.

• Law of Total Probability:

$P(A) = \sum_{i} P(A|C_i)P(C_i)$

• Bayes' Theorem:

$P(C_j|A) = \frac{P(A|C_j)P(C_j)}{\sum_{i} P(A|C_i)P(C_i)}$

Example Table: Insurance Classes

Classe	Proportion d'assurés	Probabilité de déclarer au moins un accident
1	0.15	0.09
2	0.35	0.04
3	0.34	0.06
4	0.16	0.08

Example Table: Age Groups

Age Group	Proportion of customers	Probability of reporting at least one accident
1	0.15	0.09
2	0.35	0.04
3	0.34	0.06
4	0.16	0.08

• Example Calculation: Probability that a randomly chosen customer is in group 1 given that they reported an accident:

$P(C_1|A) = \frac{P(A|C_1)P(C_1)}{P(A)} = \frac{0.09 \times 0.15}{0.0607} = 0.2224$

Normal Distribution and Quality Control

Normal Distribution Applications

The normal distribution is widely used in quality control to model measurements such as lengths or weights of manufactured items.

• Standardization: To find probabilities, standardize using:

$Z = \frac{X - \mu}{\sigma}$

• Example: If $X \sim N(9, 12.25)$, the probability that $X > 11$ is:

$P(X > 11) = P(Z > \frac{11-9}{3.5}) = P(Z > 0.57) = 1 - P(Z < 0.57) = 0.2843$

• Quality Control: To find the proportion of defective items (e.g., $X < 5.5$ or $X > 12$):

$P(5.5 < X < 12) = P(X < 12) - P(X < 5.5)$

• Variance Determination: To ensure a certain proportion of items exceed a threshold, solve for $\sigma$ using the normal table and the desired probability.

Random Variables and Probability Distributions

Discrete Random Variables

A discrete random variable can take on a finite or countable number of values, each with an associated probability.

• Probability Mass Function (PMF): For a random variable $X$, $P(X = x_i)$ gives the probability that $X$ takes value $x_i$.

• Example: If $X$ is the largest of two numbers drawn from balls labeled 1, 2, 3, the PMF is:

Result	(1,0)	(1,2)	(1,6)	(3,0)	(3,2)	(3,6)
X	1	2	6	3	2	6
P(X=x)	1/6	2/6	1/6	1/6	2/6	1/6

• Expected Value:

$E(X) = \sum x_i P(X = x_i)$ $= (1 \times 1/6) + (2 \times 2/6) + (3 \times 1/6) + (6 \times 2/6) = 3.5$

• Variance:

$Var(X) = E(X^2) - [E(X)]^2$

• Probability in an Interval: To find $P(a < X < b)$, sum the probabilities for all $x_i$ in $(a, b)$.

Key Formulas and Definitions

• Binomial Probability: $P(X = k) = \binom{n}{k} p^k (1-p)^{n-k}$

• Mean (Expected Value): $E(X) = \sum x_i P(X = x_i)$

• Variance: $Var(X) = E(X^2) - [E(X)]^2$

• Standardization (Normal): $Z = \frac{X - \mu}{\sigma}$

• Coefficient of Variation: $CV = \frac{s}{\bar{x}} \times 100$

• Law of Total Probability: $P(A) = \sum_{i} P(A|C_i)P(C_i)$

• Bayes' Theorem: $P(C_j|A) = \frac{P(A|C_j)P(C_j)}{\sum_{i} P(A|C_i)P(C_i)}$

Summary Table: Key Statistical Measures

Measure	Definition	Formula
Mean	Average value	$\bar{x} = \frac{1}{n} \sum x_i$
Variance	Average squared deviation	$s^2 = \frac{1}{n-1} \sum (x_i - \bar{x})^2$
Standard Deviation	Square root of variance	$s = \sqrt{s^2}$
Coefficient of Variation	Relative dispersion	$CV = \frac{s}{\bar{x}} \times 100$
Percentile	Value below which p% of data fall	See percentile formula above

Additional info: Some explanations and context have been expanded for clarity and completeness, including definitions, formulas, and step-by-step solutions to ensure the notes are self-contained and suitable for exam preparation.