Discrete Probability Distributions and the Binomial Distribution Study Notes

Chapter Five: Discrete Probability Distributions Overview

  • Outline of Chapter Five:
        * 5-1: Probability distributions.
        * 5-2: Mean, variance, standard deviation, and expectation.
        * 5-3: The Binomial distribution.
        * 5-4: The Binomial probability distribution shape.

  • Primary Objectives:
        * Construct a probability distribution for a random variable.
        * Calculate the mean, variance, and expected value for a discrete random variable.
        * Determine the exact probability for XX successes in nn trials of a binomial experiment.
        * Determine the mean, variance, and standard deviation for a variable within a binomial distribution.

Fundamental Definitions

  • Variable: A characteristic or attribute that can assume different values.

  • Random Variable (xx): A variable that has a single numerical value, determined by chance, for every outcome of a specific procedure.

  • Discrete Random Variable: A variable that has either a finite number of values or a countable number of values. "Countable" indicates that while there might be infinitely many values, they result from a counting process.

  • Continuous Random Variable: A variable that has infinitely many values. These values are associated with measurements on a continuous scale with no gaps or interruptions.

  • Categorization of Probability Distributions:
        * Discrete Probability Distributions: Includes specific types such as Binomial and Poisson.
        * Continuous Probability Distributions: Includes specific types such as Uniform and Normal.

Discrete Probability Distributions

  • Definition: A probability distribution consists of the values a random variable can assume and the corresponding probabilities of those values.

  • Presentation Formats: Probability distributions can be represented by a graph, a table, or a formula that provides the probability for each value of the random variable.

  • Procedure Example: Tossing Two Coins:
        * Tree Diagram Process:
            * First Toss: Head (HH) or Tail (TT).
            * Second Toss: Results in four outcomes: HH,HT,TH,TTHH, HT, TH, TT.
        * Sample Space: HH,HT,TH,TT{HH, HT, TH, TT}.
        * Random Variable (XX): If XX represents the number of heads, then XX can assume values of 0,1,extor20, 1, ext{ or } 2.
        * Distribution Table:
            * Outcome X=0X = 0: Probability P(X)=1/4P(X) = 1/4.
            * Outcome X=1X = 1: Probability P(X)=2/4P(X) = 2/4.
            * Outcome X=2X = 2: Probability P(X)=1/4P(X) = 1/4.

  • Requirements for Probability Distributions:
        * ∑P(x)=1\sum P(x) = 1 where xx assumes all possible values.
        * 0≤P(x)≤10 \le P(x) \le 1 for every individual value of xx.

Mean, Variance, and Standard Deviation

  • Mean of a Random Variable (μ\mu):
        * Formula: μ=∑[x⋅P(x)]\mu = \sum [x \cdot P(x)]

  • Variance (σ2\sigma^2):
        * Conceptual Formula: σ2=∑[(x−μ)2⋅P(x)]\sigma^2 = \sum [(x - \mu)^2 \cdot P(x)]
        * Shortcut Formula: σ2=[∑x2⋅P(x)]−μ2\sigma^2 = [\sum x^2 \cdot P(x)] - \mu^2

  • Standard Deviation (σ\sigma):
        * Defined as the square root of the variance: σ=σ2\sigma = \sqrt{\sigma^2}.

Practical Examples of Discrete Distributions

  • Example 1: Tossing a Die:
        * Distribution:
            * X∈1,2,3,4,5,6X \in {1, 2, 3, 4, 5, 6}
            * P(X)=1/6P(X) = 1/6 for each value.
        * Calculations:
            * μ=(1×1/6)+(2×1/6)+(3×1/6)+(4×1/6)+(5×1/6)+(6×1/6)=21/6=3.5\mu = (1 \times 1/6) + (2 \times 1/6) + (3 \times 1/6) + (4 \times 1/6) + (5 \times 1/6) + (6 \times 1/6) = 21/6 = 3.5
            * ∑X2⋅P(X)=12(1/6)+22(1/6)+32(1/6)+42(1/6)+52(1/6)+62(1/6)=91/6≈15.167\sum X^2 \cdot P(X) = 1^2(1/6) + 2^2(1/6) + 3^2(1/6) + 4^2(1/6) + 5^2(1/6) + 6^2(1/6) = 91/6 \approx 15.167
            * σ2=15.167−(3.5)2=15.167−12.25=2.917\sigma^2 = 15.167 - (3.5)^2 = 15.167 - 12.25 = 2.917
            * σ=2.917≈1.71\sigma = \sqrt{2.917} \approx 1.71

  • Example 2: Radio Talk Show Phone Lines:
        * Problem: Find the standard deviation for the number of people placed on hold (0 to 4 lines).
        * Distribution Data:
            * P(0)=0.18P(0) = 0.18
            * P(1)=0.34P(1) = 0.34
            * P(2)=0.23P(2) = 0.23
            * P(3)=0.21P(3) = 0.21
            * P(4)=0.04P(4) = 0.04
        * Calculations:
            * μ=(0)(0.18)+(1)(0.34)+(2)(0.23)+(3)(0.21)+(4)(0.04)=1.59\mu = (0)(0.18) + (1)(0.34) + (2)(0.23) + (3)(0.21) + (4)(0.04) = 1.59
            * ∑X2P(X)=(02)(0.18)+(12)(0.34)+(22)(0.23)+(32)(0.21)+(42)(0.04)=3.79\sum X^2 P(X) = (0^2)(0.18) + (1^2)(0.34) + (2^2)(0.23) + (3^2)(0.21) + (4^2)(0.04) = 3.79
            * μ2=1.592=2.5281≈2.53\mu^2 = 1.59^2 = 2.5281 \approx 2.53
            * σ2=3.79−2.53=1.26\sigma^2 = 3.79 - 2.53 = 1.26
            * σ=1.26≈1.12\sigma = \sqrt{1.26} \approx 1.12

Mathematical Expectation

  • Expected Value (E(X)E(X)): The expected value of a discrete random variable is the theoretical average of the variable. It represents the mean of a probability distribution.

  • Formula: E(X)=μ=∑[X⋅P(X)]E(X) = \mu = \sum [X \cdot P(X)]

  • Case Study: Ski Resort Profit:
        * Scenario: A resort loses $70,000\$70,000 if it does not snow much (Probability P=0.60P = 0.60) and makes $250,000\$250,000 if it snows at least 75 inches75\,\text{inches} (Probability P=0.40P = 0.40).
        * Distribution Table:
            * Profit (X=250,000X = 250,000), P(X)=0.40P(X) = 0.40
            * Profit (X=−70,000X = -70,000), P(X)=0.60P(X) = 0.60
        * Calculation:
            * E(X)=(250,000)(0.40)+(−70,000)(0.60)=100,000−42,000=$58,000E(X) = (250,000)(0.40) + (-70,000)(0.60) = 100,000 - 42,000 = \$58,000
        * Conclusion: The expected profit for the season is $58,000\$58,000.

The Binomial Distribution

  • Definition: A binomial experiment is a probability experiment that must satisfy four specific requirements:
        1. Each trial has only two outcomes (or outcomes reducible to two), categorized as "success" or "failure."
        2. The experiment has a fixed number of trials (nn).
        3. The trials must be independent; the outcome of one trial does not affect the probability of others.
        4. The probability of success (pp) must remain constant across all trials.

  • Notation:
        * P(Success)=pP(\text{Success}) = p
        * P(Failure)=1−p=qP(\text{Failure}) = 1 - p = q
        * nn = number of trials
        * XX = number of successes among nn trials.

  • Binomial Distribution Formulas:
        * Probability of exactly XX successes: P(X)=(nX)pXqn−XP(X) = \binom{n}{X} p^X q^{n-X}
        * Mean: μ=np\mu = np
        * Variance: σ2=npq\sigma^2 = npq
        * Standard Deviation: σ=npq\sigma = \sqrt{npq}

Binomial Distribution Content and Contextual Note

  • Clarification on "Success": In the context of probability, "success" does not imply a positive or good outcome; it merely refers to the occurrence of the specific outcome being studied.
        * Example: If selecting individuals to see who has been in an auto accident, an accident would be defined as a "success."

Practical Exercises for Binomial Distributions

  • Exercise A: Guessing on Multiple-Choice Questions:
        * Scenario: Student guesses on 55 questions with 55 choices each (p=1/5=0.2p = 1/5 = 0.2).
        * Goal: Find the probability of getting exactly 33 correct.
        * Parameters: n=5n = 5, X=3X = 3, p=0.2p = 0.2.
        * Calculation: P(3)=5!(5−3)!3!×(0.2)3×(0.8)2≈0.0512P(3) = \frac{5!}{(5 - 3)!3!} \times (0.2)^3 \times (0.8)^2 \approx 0.0512.

  • Exercise B: Teenagers and Part-time Jobs:
        * Scenario: 30%30\% of teenagers (p=0.3p=0.3) have part-time jobs. Sample size n=15n=15.
        * 1. Find exactly 4 teens with jobs:
            * P(X=4)=(154)(0.3)4(0.7)11≈0.22P(X=4) = \binom{15}{4} (0.3)^4 (0.7)^{11} \approx 0.22
        * 2. At least 2 teens with jobs:
            * P(X≥2)=1−[P(X=0)+P(X=1)]P(X \ge 2) = 1 - [P(X=0) + P(X=1)]
            * P(X=0)=(150)(0.3)0(0.7)15≈0.0047P(X=0) = \binom{15}{0} (0.3)^0 (0.7)^{15} \approx 0.0047
            * P(X=1)=(151)(0.3)1(0.7)14≈0.0305P(X=1) = \binom{15}{1} (0.3)^1 (0.7)^{14} \approx 0.0305
            * P(X≥2)=1−(0.0047+0.0305)=0.9648≈0.96P(X \ge 2) = 1 - (0.0047 + 0.0305) = 0.9648 \approx 0.96
        * 3. At most 13 do not have part-time jobs:
            * Let YY be the number without jobs (p=0.7p=0.7 for success here).
            * P(Y≤13)=1−[P(Y=14)+P(Y=15)]P(Y \le 13) = 1 - [P(Y=14) + P(Y=15)]
            * P(Y=14)=(1514)(0.7)14(0.3)1=0.0305P(Y=14) = \binom{15}{14} (0.7)^{14} (0.3)^1 = 0.0305
            * P(Y=15)=(1515)(0.7)15(0.3)0=0.0047P(Y=15) = \binom{15}{15} (0.7)^{15} (0.3)^0 = 0.0047 (approx: transcript lists 0.000000040.00000004 likely as a typo for the decimal place or extremely small value).
            * P(Y≤13)=1−(0.0305+4.74imes10−3)≈0.9648P(Y \le 13) = 1 - (0.0305 + 4.74 imes 10^{-3}) \approx 0.9648
        * 4. Average and Standard Deviation (for teens with jobs):
            * μ=np=15×0.3=4.5\mu = np = 15 \times 0.3 = 4.5
            * σ2=npq=15×0.3×0.7=3.15\sigma^2 = npq = 15 \times 0.3 \times 0.7 = 3.15
            * σ=3.15≈1.77\sigma = \sqrt{3.15} \approx 1.77

  • Example: Tossing a Coin Four Times:
        * Find mean and variance for number of heads.
        * n=4,p=0.5,q=0.5n = 4, p = 0.5, q = 0.5
        * μ=4×0.5=2\mu = 4 \times 0.5 = 2
        * σ2=4×0.5×0.5=1\sigma^2 = 4 \times 0.5 \times 0.5 = 1
        * σ=1=1\sigma = \sqrt{1} = 1

  • Example: Random Guessing on 10-Question Quiz:
        * Conditions: n=10,p=0.2n = 10, p = 0.2. Find P(X=6)P(X=6).
        * P(6)=(106)(0.2)6(0.8)4=210×0.000064×0.4096=0.005505P(6) = \binom{10}{6} (0.2)^6 (0.8)^4 = 210 \times 0.000064 \times 0.4096 = 0.005505.
        * Conclusion: Chance is less than 1%1\%. The chance of passing (X≥6X \ge 6) is also less than 1%1\%.

Distribution Shapes and Properties

  • Histogram Behaviors based on pp:
        * If p<0.5p < 0.5, the distribution histogram is skewed right (e.g., n=10,p=0.2n=10, p=0.2; Mean =2= 2).     * If p=0.5p = 0.5, the distribution histogram is symmetric (e.g., n=10,p=0.5n=10, p=0.5; Mean =5= 5).     * If p>0.5p > 0.5, the distribution histogram is skewed left (e.g., n=10,p=0.8n=10, p=0.8; Mean =8= 8).

  • The Effect of Sample Size (nn):
        * Even if the distribution is initially skewed, the histograms become increasingly bell-shaped as the value of nn increases.
        * Example Visualization: Comparing histograms for n=30,p=0.2n=30, p=0.2 versus n=70,p=0.2n=70, p=0.2 shows the trend toward a normal shape.