Discrete Probability Distributions and Random Variables
Foundation of Random Variables
- A random experiment generates a sample space , which represents the set of all possible outcomes.
- A random variable (also termed a chance variable or stochastic variable), denoted by , is a rule that assigns a unique real number to each outcome .

- Coin Tossing Example:
- Consider an experiment where a fair coin is tossed times.
- The sample space consists of elementary outcomes:
- Let denote the number of heads obtained in the three tosses.
- Mapping of sample points to real line values:
- The random variable takes values in the set .
- Associated probabilities:

Classification of Random Variables
- Random variables are broadly classified into two categories:
- Discrete Random Variable (DRV): Has a finite or countably infinite number of possible outcomes that can be listed. Represents count data in practical applications.
- Example: Number of defectives in a sample.
- Example: The number of automobile accidents per year in Whitefield.
- Example: The number of eggs laid each month by a hen.
- Continuous Random Variable (CRV): Takes an uncountable number of values across a continuous interval. Represents measured data in practical applications.
- Example: Heights, weights, distance, temperature.
- Example: The length of time to play holes of golf.
- Example: The amount of milk produced by a particular cow.
Discrete Probability Functions and Parameter Formulas
- For a discrete random variable , each distinct value is assigned a real number probability .
- Axiomatic Requirements for Discrete Probability Density Function (PDF) / Probability Mass Function:
- Non-negativity and upper bound: for every value x$.\n 2. Completeness (Sum of probabilities): \sum P(x) = 1.\n\n* **Key Measures of a Discrete Probability Distribution**:\n * **Mean (Expected Value)**:\n \mu = E(X) = \sum x P(x)\n * **Variance**:\n \sigma^2 = Var(X) = \sum (x - \mu)^2 P(x)\n * Computational alternative formula:\n \sigma^2 = E[X^2] - [E(X)]^2 = \sum x^2 P(x) - \mu^2\n where E[X^2] = \sum x^2 P(x).\n * **Standard Deviation**:\n \sigma = \sqrt{Var(X)} = \sqrt{\sigma^2}\n\n# Detailed Worked Examples on Discrete Random Variables\n\n* **Example 1: Rolling Two Six-Sided Dice**\n * Two fair six-sided dice are rolled. Total possible outcomes = 36.\n * Let XX \in {2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}).\n * Probability distribution and calculation details:\n * x = 2: outcomes \{(1,1)\login\}, P(2) = \frac{1}{36}x P(x) = \frac{2}{36}x^2 P(x) = \frac{4}{36}\n * x = 3: outcomes \{(1,2), (2,1)\login\}, P(3) = \frac{2}{36}x P(x) = \frac{6}{36}x^2 P(x) = \frac{18}{36}\n * x = 4: outcomes \{(1,3), (2,2), (3,1)\login\}, P(4) = \frac{3}{36}x P(x) = \frac{12}{36}x^2 P(x) = \frac{48}{36}\n * x = 5: outcomes \{(1,4), (2,3), (3,2), (4,1)\login\}, P(5) = \frac{4}{36}x P(x) = \frac{20}{36}x^2 P(x) = \frac{100}{36}\n * x = 6: outcomes \{(1,5), (2,4), (3,3), (4,2), (5,1)\login\}, P(6) = \frac{5}{36}x P(x) = \frac{30}{36}x^2 P(x) = \frac{180}{36}\n * x = 7: outcomes \{(1,6), (2,5), (3,4), (4,3), (5,2), (6,1)\login\}, P(7) = \frac{6}{36}x P(x) = \frac{42}{36}x^2 P(x) = \frac{294}{36}\n * x = 8: outcomes \{(2,6), (3,5), (4,4), (5,3), (6,2)\login\}, P(8) = \frac{5}{36}x P(x) = \frac{40}{36}x^2 P(x) = \frac{320}{36}\n * x = 9: outcomes \{(3,6), (4,5), (5,4), (6,3)\login\}, P(9) = \frac{4}{36}x P(x) = \frac{36}{36}x^2 P(x) = \frac{324}{36}\n * x = 10: outcomes \{(4,6), (5,5), (6,4)\login\}, P(10) = \frac{3}{36}x P(x) = \frac{30}{36}x^2 P(x) = \frac{300}{36}\n * x = 11: outcomes \{(5,6), (6,5)\login\}, P(11) = \frac{2}{36}x P(x) = \frac{22}{36}x^2 P(x) = \frac{242}{36}\n * x = 12: outcomes \{(6,6)\login\}, P(12) = \frac{1}{36}x P(x) = \frac{12}{36}x^2 P(x) = \frac{144}{36}\n * Totals:\n * \sum P(x) = \frac{36}{36} = 1.00\n * \mu = E(X) = \sum x P(x) = \frac{252}{36} = 7.0\n * E[X^2] = \sum x^2 P(x) = \frac{1974}{36} \approx 54.8333\n * Variance and Standard Deviation:\n * \sigma^2 = E[X^2] - \mu^2 = 54.8333 - (7.0)^2 = 54.8333 - 49 = 5.8333\n * \sigma = \sqrt{5.8333} = 2.4152\n\n* **Example 2: Verification of Distribution Validity and Parameter Evaluation**\n * Given distribution:\n * P(10) = \frac{1}{8}\n * P(20) = \frac{3}{8}\n * P(30) = \frac{3}{8}\n * P(40) = \frac{1}{8}\n * Verification:\n * 0 \le P(x) \le 1x$.
- \sum P(x) = \frac{1}{8} + \frac{3}{8} + \frac{3}{8} + \frac{1}{8} = \frac{8}{8} = 1$.\n * The distribution represents a valid discrete probability function.\n * Mean calculation:\n * \mu = \sum x P(x) = 10\left(\frac{1}{8}\right) + 20\left(\frac{3}{8}\right) + 30\left(\frac{3}{8}\right) + 40\left(\frac{1}{8}\right) = \frac{10 + 60 + 90 + 40}{8} = \frac{200}{8} = 25\n * Variance calculation:\n * \sigma^2 = \sum (x - \mu)^2 P(x) = (10-25)^2\left(\frac{1}{8}\right) + (20-25)^2\left(\frac{3}{8}\right) + (30-25)^2\left(\frac{3}{8}\right) + (40-25)^2\left(\frac{1}{8}\right)\n * \sigma^2 = \frac{225}{8} + \frac{75}{8} + \frac{75}{8} + \frac{225}{8} = \frac{600}{8} = 75\n * Standard deviation:\n * \sigma = \sqrt{75}\n\n* **Example 3: Ship Arrival Probabilities at Harbor**\n * Probability distribution for number of ships x arriving daily:\n * P(10) = 0.4\n * P(11) = 0.2\n * P(12) = 0.2\n * P(13) = 0.1\n * P(14) = 0.1\n * Questions:\n * a. Probability that exactly 14 ships arrive:\n P(X = 14) = 0.1\n * b. Probability that at least 12 ships arrive:\n P(X \ge 12) = P(12) + P(13) + P(14) = 0.2 + 0.1 + 0.1 = 0.4\n * c. Probability that at most 11 ships arrive:\n P(X \le 11) = P(10) + P(11) = 0.4 + 0.2 = 0.6\n\n* **Example 4: Solving for Constant k and Interval Probabilities**\n * Given probability function for x \in {-3, -2, -1, 0, 1, 2, 3}:\n * P(-3) = kP(-2) = 2kP(-1) = 3kP(0) = 4kP(1) = 3kP(2) = 2kP(3) = k\n * Step 1: Solve for k:\n \sum P(x) = k + 2k + 3k + 4k + 3k + 2k + k = 16k = 1 \implies k = \frac{1}{16}\n * Distribution table with k = \frac{1}{16}:\n * P(-3) = \frac{1}{16}P(-2) = \frac{2}{16}P(-1) = \frac{3}{16}P(0) = \frac{4}{16}P(1) = \frac{3}{16}P(2) = \frac{2}{16}P(3) = \frac{1}{16}\n * Mean calculation:\n \mu = \sum x P(x) = -3\left(\frac{1}{16}\right) - 2\left(\frac{2}{16}\right) - 1\left(\frac{3}{16}\right) + 0\left(\frac{4}{16}\right) + 1\left(\frac{3}{16}\right) + 2\left(\frac{2}{16}\right) + 3\left(\frac{1}{16}\right) = \frac{0}{16} = 0\n * Variance calculation:\n * E[X^2] = \sum x^2 P(x) = (-3)^2\left(\frac{1}{16}\right) + (-2)^2\left(\frac{2}{16}\right) + (-1)^2\left(\frac{3}{16}\right) + 0^2\left(\frac{4}{16}\right) + 1^2\left(\frac{3}{16}\right) + 2^2\left(\frac{2}{16}\right) + 3^2\left(\frac{1}{16}\right)\n * E[X^2] = \frac{9 + 8 + 3 + 0 + 3 + 8 + 9}{16} = \frac{40}{16} = \frac{5}{2}\n * \sigma^2 = E[X^2] - \mu^2 = \frac{5}{2} - 0^2 = \frac{5}{2}\n * Standard deviation:\n \sigma = \sqrt{\frac{5}{2}}\n * Interval probabilities:\n * P(X \le 1) = P(-3) + P(-2) + P(-1) + P(0) + P(1) = \frac{1 + 2 + 3 + 4 + 3}{16} = \frac{13}{16}\n * P(X > 1) = P(2) + P(3) = \frac{2 + 1}{16} = \frac{3}{16}\n * P(-1 < X \le 2) = P(0) + P(1) + P(2) = \frac{4 + 3 + 2}{16} = \frac{9}{16}\n\n* **Example 5: Testing Functions for Probability Distribution Rules**\n * Function 1: f(x) = \frac{x}{15}x \in {0, 1, 2, 3, 4, 5}\n * Non-negative: Yes (f(x) \ge 0x).\n * Sum of probabilities: \sum f(x) = \frac{0 + 1 + 2 + 3 + 4 + 5}{15} = \frac{15}{15} = 1$.
- Result: Valid probability distribution.
- Function 2: for
- Non-negative check: For , f(3) = \frac{5 - 6}{6} = -\frac{1}{6} < 0$.\n * Result: Not a valid probability distribution (violates non-negativity condition).\n * Function 3: f(x) = \frac{1}{4}x \in {3, 4, 5, 6}\n * Non-negative: Yes.\n * Sum of probabilities: \sum f(x) = \frac{1}{4} + \frac{1}{4} + \frac{1}{4} + \frac{1}{4} = 1$.
- Result: Valid probability distribution.
- Function 4: for
- Sum of probabilities: \sum f(x) = \frac{2 + 3 + 4 + 5 + 6}{25} = \frac{20}{25} = 0.8 \neq 1$.\n * Result: Not a valid probability distribution (sum does not equal 1).\n\n* **Example 6: Sampling Defective Items from a Box**\n * Box contents: Total 1248 good).\n * Sample size: 3XX \in {0, 1, 2, 3}).\n * Distribution calculations:\n * P(X = 0) = \frac{\binom{4}{0} \times \binom{8}{3}}{\binom{12}{3}} = \frac{1 \times 56}{220} = \frac{14}{55}\n * P(X = 1) = \frac{\binom{4}{1} \times \binom{8}{2}}{\binom{12}{3}} = \frac{4 \times 28}{220} = \frac{28}{55}\n * P(X = 2) = \frac{\binom{4}{2} \times \binom{8}{1}}{\binom{12}{3}} = \frac{6 \times 8}{220} = \frac{12}{55}\n * P(X = 3) = \frac{\binom{4}{3} \times \binom{8}{0}}{\binom{12}{3}} = \frac{4 \times 1}{220} = \frac{1}{55}\n * Mean calculation:\n \mu = \sum x P(x) = 0\left(\frac{14}{55}\right) + 1\left(\frac{28}{55}\right) + 2\left(\frac{12}{55}\right) + 3\left(\frac{1}{55}\right) = \frac{28 + 24 + 3}{55} = \frac{55}{55} = 1.0\n * Standard deviation calculation:\n * E[X^2] = 0^2\left(\frac{14}{55}\right) + 1^2\left(\frac{28}{55}\right) + 2^2\left(\frac{12}{55}\right) + 3^2\left(\frac{1}{55}\right) = \frac{28 + 48 + 9}{55} = \frac{85}{55} = \frac{17}{11}\n * \sigma^2 = E[X^2] - \mu^2 = \frac{17}{11} - (1.0)^2 = \frac{6}{11}\n * \sigma = \sqrt{\frac{6}{11}}\n\n* **Example 7: Probability Distribution with Parameter K**\n * Distribution for x \in {-2, -1, 0, 1, 2, 3}:\n * P(-2) = 0.1P(-1) = KP(0) = 0.2P(1) = 2KP(2) = 0.3P(3) = K\n * Solve for K:\n \sum P(x) = 0.1 + K + 0.2 + 2K + 0.3 + K = 0.6 + 4K = 1 \implies 4K = 0.4 \implies K = 0.1\n * Evaluated probabilities (K = 0.1):\n * P(-2) = 0.1P(-1) = 0.1P(0) = 0.2P(1) = 0.2P(2) = 0.3P(3) = 0.1\n * Mean:\n \mu = -2(0.1) - 1(0.1) + 0(0.2) + 1(0.2) + 2(0.3) + 3(0.1) = -0.2 - 0.1 + 0 + 0.2 + 0.6 + 0.3 = 0.8\n * Variance:\n * E[X^2] = (-2)^2(0.1) + (-1)^2(0.1) + 0^2(0.2) + 1^2(0.2) + 2^2(0.3) + 3^2(0.1) = 0.4 + 0.1 + 0 + 0.2 + 1.2 + 0.9 = 2.8\n * \sigma^2 = E[X^2] - \mu^2 = 2.8 - (0.8)^2 = 2.8 - 0.64 = 2.16\n * \sigma = \sqrt{2.16}\n * Tail and range probabilities:\n * P(X < 1) = P(-2) + P(-1) + P(0) = 0.1 + 0.1 + 0.2 = 0.4\n * P(X > -1) = P(0) + P(1) + P(2) + P(3) = 0.2 + 0.2 + 0.3 + 0.1 = 0.8\n * P(-1 < X \le 2) = P(0) + P(1) + P(2) = 0.2 + 0.2 + 0.3 = 0.7\n\n* **Example 8: Quadratic Parameter Equation in Discrete RV**\n * Probability function for x \in {0, 1, 2, 3, 4, 5, 6, 7}:\n * P(0) = 0P(1) = KP(2) = 2KP(3) = 2KP(4) = 3KP(5) = K^2P(6) = 2K^2P(7) = 7K^2 + K\n * Step 1: Solve for K:\n \sum P(x) = 0 + K + 2K + 2K + 3K + K^2 + 2K^2 + (7K^2 + K) = 10K^2 + 9K = 1\n 10K^2 + 9K - 1 = 0 \implies (10K - 1)(K + 1) = 0\n Since P(x) \ge 0K = \frac{1}{10} = 0.1$.
- Evaluated probabilities:
- Probability evaluations:
Bernoulli Trials and the Binomial Distribution
Bernoulli Trials: A sequence of trials is defined as Bernoulli trials if:
- Each trial has only two possible outcomes: Success () with probability , and Failure () with probability q = 1 - p$.\n 2. The probability of success p remains identical across all trials.\n 3. The trials are strictly independent.\n * Example: Rolling a fair die 106$.
- (probability of getting ).
- (probability of getting ).
- Outcomes of separate rolls are independent and remains constant.
Binomial Experiment:
- Consists of repeated independent trials.
- Each trial yields two outcomes ( or ).
- Probability of success remains constant.
- Random variable represents the count of successes in trials ().
Mathematical Translation Table for Inequalities:

Binomial Probability Formula: where:
- = total number of trials
- = probability of success on a single trial
- = probability of failure on a single trial
- = exact number of successes ()
Population Parameters of a Binomial Distribution:
- Mean:
- Variance:
- Standard Deviation:
Detailed Binomial Examples:
Card Experiment:
Randomly draw a card from a standard deck of cards, record if it is an Ace, replace the card, and repeat times.
Evaluation: Binomial experiment because trials are independent due to replacement, there are two outcomes (Ace or non-Ace), and probability remains fixed.
Parameters: , , , .
Red Chips Experiment 1 (Sample Size ):
A bag contains chips ( red, white, blue). Three chips are selected with replacement ().
Success = selecting a red chip , q = 0.7$.\n * Probability of selecting exactly one red chip (x = 1):\n P(1) = {}_3C_1 (0.3)^1 (0.7)^2 = 3 \times 0.3 \times 0.49 = 0.441\n\n * **Red Chips Experiment 2 (Sample Size n = 4)**:\n * Bag has 10352n = 4).\n * Parameters: p = 0.3q = 0.7$.
Probability Distribution Table:
Probability Calculations:
- a. Probability of selecting no more than red chips () :
- b. Probability of selecting at least red chip () using complement rule:
College Breakfast Skipping Example:
Proportion of students skipping breakfast = out of (, ). Sample size n = 10$.\n * Mean: \mu = n p = 10 \times 0.2 = 2.0\n * Variance: \sigma^2 = n p q = 10 \times 0.2 \times 0.8 = 1.6\n * Standard Deviation: \sigma = \sqrt{1.6} \approx 1.2649\n\n * **Family Children Demographics Example**:\n * Total families N = 800n = 5p = \frac{1}{2}q = \frac{1}{2}x be the number of boys.\n * i. Expected number of families with 3x = 3):\n P(X = 3) = {}_5C_3 \left(\frac{1}{2}\right)^3 \left(\frac{1}{2}\right)^2 = 10 \times \frac{1}{32} = \frac{10}{32}\n \text{Expected count} = 800 \times \frac{10}{32} = 250\n * ii. Expected number of families with 5x = 0 boys):\n P(X = 0) = {}_5C_0 \left(\frac{1}{2}\right)^0 \left(\frac{1}{2}\right)^5 = 1 \times \frac{1}{32} = \frac{1}{32}\n \text{Expected count} = 800 \times \frac{1}{32} = 25\n * iii. Expected number of families with either 23 boys:\n P(X = 2 \text{ or } 3) = P(2) + P(3) = \frac{10}{32} + \frac{10}{32} = \frac{20}{32}\n \text{Expected count} = 800 \times \frac{20}{32} = 500\n * iv. Expected number of families with at most 23, 4, \text{ or } 5\implies X \ge 3):\n P(X \ge 3) = P(3) + P(4) + P(5) = \left[{}_5C_3 + {}_5C_4 + {}_5C_5\right] \left(\frac{1}{2}\right)^5 = \frac{10 + 5 + 1}{32} = \frac{16}{32}\n \text{Expected count} = 800 \times \frac{16}{32} = 400\n\n * **Telephone Line Busy Example**:\n * Probability of busy line on weekday between 2 pm and 3 pm = \frac{1}{15}p = \frac{1}{15}q = \frac{14}{15}n = 6$.
i. Probability that not more than lines are busy () :
ii. Probability that at least lines are busy () :
Hypergeometric Distribution
Characteristics:
- A sample of elements is randomly selected without replacement from a finite population of total elements.
- The population contains elements categorized as Successes () and elements categorized as Failures ().
- The random variable measures the exact number of successes obtained in the sample of size n$.\n 4. Trials are dependent because sampling occurs without replacement.\n\n* **Hypergeometric Probability Formula**:\n P(x) = \frac{\binom{X}{x} \binom{N - X}{n - x}}{\binom{N}{n}}\n where combination operator is defined as:\n \binom{N}{x} = \frac{N!}{x!(N - x)!}\n * Definitions of variables:\n * N = Total number of items in population (possible trials)\n * n = Sample size drawn (actual trials)\n * X = Total number of successes available in population (possible successes)\n * x = Number of successes obtained in sample (actual successes)\n\n* **Detailed Hypergeometric Examples**:\n\n * **CD Batch Inspection Example**:\n * Shipment size N = 1000X = 20N - X = 980$.
Inspection sample size n = 30$.\n * Rule: Retailer returns batch if 3x \ge 3).\n * Calculation via complement rule:\n P(\text{return shipment}) = 1 - [P(0) + P(1) + P(2)]\n * P(0) = \frac{\binom{20}{0}\binom{1000 - 20}{30 - 0}}{\binom{1000}{30}} = \frac{\binom{20}{0}\binom{980}{30}}{\binom{1000}{30}} = 0.541\n * P(1) = \frac{\binom{20}{1}\binom{1000 - 20}{30 - 1}}{\binom{1000}{30}} = \frac{\binom{20}{1}\binom{980}{29}}{\binom{1000}{30}} = 0.341\n * P(2) = \frac{\binom{20}{2}\binom{1000 - 20}{30 - 2}}{\binom{1000}{30}} = \frac{\binom{20}{2}\binom{980}{28}}{\binom{1000}{30}} = 0.099\n * P(\text{return shipment}) = 1 - (0.541 + 0.341 + 0.099) = 1 - 0.981 = 0.019 = 1.9\%\n\n * **Light Bulb Carton Example**:\n * Carton of N = 24X = 3N - X = 21n = 6$.
Probability function formula:
Probability of selecting defective bulbs ():
Probability of selecting all defective bulbs ():
Automobile Safety Inspection Example:
Lot size , cars inspected . Below safety standards , meeting standards N - X = 8$.\n * Find probability that at least 1P(X \ge 1) = P(1) + P(2)).\n * P(1) = \frac{\binom{2}{1}\binom{10 - 2}{5 - 1}}{\binom{10}{5}} = \frac{\binom{2}{1}\binom{8}{4}}{\binom{10}{5}} = \frac{\frac{2!}{1!1!} \times \frac{8!}{4!4!}}{\frac{10!}{5!5!}} = \frac{2 \times 70}{252} = \frac{5}{9} \approx 0.556\n * P(2) = \frac{\binom{2}{2}\binom{10 - 2}{5 - 2}}{\binom{10}{5}} = \frac{\binom{2}{2}\binom{8}{3}}{\binom{10}{5}} = \frac{\frac{2!}{2!0!} \times \frac{8!}{3!5!}}{\frac{10!}{5!5!}} = \frac{1 \times 56}{252} = \frac{2}{9} \approx 0.222\n * Total Probability P(1) + P(2) = \frac{5}{9} + \frac{2}{9} = \frac{7}{9} \approx 0.778$$