Discrete and Continuous Random Variables and Probability Distributions

Fundamentals of Random Variables

• Classification of Variables:
  • Qualitative Variables: Categorical attributes or descriptive qualities.
  • Quantitative Variables: Numerical values that can be measured or counted. Quantitative variables are subdivided into:
    • Discrete Variables: Variables whose possible values are whole numbers or countable outcomes.
    • Continuous Variables: Variables whose possible values can take any numerical value within an interval, including fractions and decimals.
• Definition of a Random Variable:
  • A random variable is a variable whose value is determined by chance or probability in a random experiment.
  • While standard mathematical variables have defined fixed values (e.g., $x = 0$), a random variable associates each possible numerical outcome with a specific probability or chance of occurrence (e.g., $P(X = 0) = \frac{1}{2}$ and $P(X = 1) = \frac{1}{2}$).
• Notation Rules:
  • Capital/Uppercase Letters ($X$, $Y$, $Z$): Represent the random variable itself, containing the set of all possible outcomes/values.
  • Lowercase Letters ($x$, $y$, $z$): Represent a specific individual numerical value that the random variable can take (e.g., $x = 0$ or $x = 1$).
  • Probability Notation: $P(X = x)$ or $P(x)$ denotes the probability that the random variable $X$ equals the specific value $x$
• Statistical Context:
  • Discrete probability distributions (covered in Chapter 5) and continuous probability distributions (covered in Chapter 6) serve as essential foundational concepts for inferential statistics (statistical inference) presented in Chapters 8, 9, and 10.

Discrete vs. Continuous Random Variables

• Discrete Random Variables:
  • Definition: A random variable that has either a finite or countable number of possible values.
  • Characteristics: Values are distinct, separate whole numbers. Fractions or intermediate decimal values between possible outcomes are impossible.
  • Examples & Counter-Examples:
    • Coin Flipping Example: Let $X$ be the number of heads in two coin tosses.
      • Sample Space $S = \{HH, HT, TH, TT\}$.
      • Outcome mapping to $X$:
        • $HH \rightarrow 2$
        • $HT \rightarrow 1$
        • $TH \rightarrow 1$
        • $TT \rightarrow 0$
      • Possible values of $X$ are $0$, $1$, and $2$. (A value such as $1.5$ heads is impossible).
    • Patient Recovery Example: Let $Y$ be the number of patients recovering after a specific medical treatment.
      • Possible values for $Y$ are $0$, $1$, $2$, $3$, $\dots$
      • Values represent whole counts of individual people.
    • M&M Color Count: Let $X$ be the number of distinct colors of M&M's in a bag.
      • Classification: Discrete.
      • Possible values: $0$, $1$, $2$, $3$, $\dots$
• Continuous Random Variables:
  • Definition: A random variable that has infinitely many possible values within a continuous interval.
  • Characteristics: Values are measurement-based and can take any decimal or fractional real number within a range.
  • Examples:
    • Student Height: Let $X$ be the height of students in inches.
      • Possible values can be any real number: $63\,\text{in}$, $63.5\,\text{in}$, $63.55\,\text{in}$, etc.
      • There are infinitely many possible values depending on precision.
    • Task Completion Time: Let $Y$ be the time taken to complete a task in hours.
      • Possible values can be any continuous measure: $2\,\text{hours}$, $2.5\,\text{hours}$, $2.75\,\text{hours}$, etc.
    • M&M Mass: Mass of an individual M&M.
      • Classification: Continuous.
      • Possible values: $1\,\text{g}$, $2.5\,\text{g}$, etc. (continuous measurement).
    • M&M Diameter: Diameter of an individual M&M.
      • Classification: Continuous.
      • Possible values: $1\,\text{mm}$, $1.1\,\text{mm}$, $2.75\,\text{mm}$, etc. (continuous measurement).

Discrete Probability Distributions

• Definition:
  • The probability distribution of a discrete random variable $X$ provides all possible values of $X$ alongside their corresponding probabilities $P(x)$.
• Methods of Representation:
  • Tables
  • Graphs (bar-style probability graphs)
  • Formulas
• Two Essential Validation Rules:
  1. Probability Range Rule: Every individual probability $P(x)$ must be between $0$ and $1$ inclusive:         $0 \le P(x) \le 1$
  2. Summation Rule: The sum of all probabilities for all possible values of $X$ must equal exactly $1$:         $\sum P(x) = 1$
• Step-by-Step Construction Procedure:
  • Step 1: Identify all possible numerical values $x$ for the random variable $X$ using the sample space.
  • Step 2: Calculate the individual probability $P(X = x)$ for each outcome $x$.
  • Step 3: Construct a probability distribution table listing $x$ in the first column/row and $P(X = x)$ in the second column/row, and verify that the sum of probabilities equals $1$
• Coin Toss Distribution Construction Example:
  • Experiment: Flipping two fair coins; $X$ = number of heads.
  • Sample Space: $S = \{HH, HT, TH, TT\}$.
  • Probabilities:
    • $P(X = 0) = P(TT) = \frac{1}{4} = 0.25$
    • $P(X = 1) = P(HT \text{ or } TH) = \frac{1}{4} + \frac{1}{4} = \frac{2}{4} = 0.50$
    • $P(X = 2) = P(HH) = \frac{1}{4} = 0.25$
  • Distribution Table:
    • $x = 0 \implies P(x) = \frac{1}{4}$
    • $x = 1 \implies P(x) = \frac{2}{4}$
    • $x = 2 \implies P(x) = \frac{1}{4}$
  • Verification Check: $\frac{1}{4} + \frac{2}{4} + \frac{1}{4} = \frac{4}{4} = 1.00$. The distribution is valid.

Real-World Applications and Models

• Insurance Policy Profit Model:

  • Scenario Parameters:
    • Target demographic: Healthy $24$-year-old individual.
    • Probability of death within the year: $P(\text{death}) = 0.0014$
    • Insurance policy payout to beneficiaries upon death: $\$10{,}000$
    • Annual policy fee (cost paid by customer): $\$100$
    • Random Variable $X$: Amount of annual profit earned by the insurance company per policy.
  • Outcomes & Calculation:
    1. Policyholder Survives:
       • Probability: $P(\text{survive}) = 1 - 0.0014 = 0.9986$
       • Company Profit $x$: $\$100$
    2. Policyholder Dies:
       • Probability: $P(\text{death}) = 0.0014$
       • Company Profit $x$: $\$100 - \$10{,}000 = -\$9{,}900$
  • Probability Distribution Table:
    • $x = 100 \implies P(x) = 0.9986$
    • $x = -9900 \implies P(x) = 0.0014$
  • Validation Check:
    • Both $0.9986$ and $0.0014$ lie strictly within $[0, 1]$.
    • $\sum P(x) = 0.9986 + 0.0014 = 1.0000$. The distribution is valid.

• Chuck-a-Luck Casino Game Model:

  • Game Rules:
    • Three fair dice are rolled in a wire cage.
    • A player places a $\$1$ bet on a selected number from $1$ to $6$.
    • If $0$ dice match the selected number, the player loses the $\$1$ bet.
    • If $1$ die matches, the player wins $\$1$ (net profit $\$1$).
    • If $2$ dice match, the player wins $\$2$ (net profit $\$2$).
    • If $3$ dice match, the player wins $\$3$ (net profit $\$3$).
  • Random Variable $X$: Profit from a $\$1$ bet in Chuck-a-luck.
  • Given Probability Distribution Table:
    • $0$ Matches: $x = -1 \implies P(x) = 0.5787$
    • $1$ Match: $x = 1 \implies P(x) = 0.3472$
    • $2$ Matches: $x = 2 \implies P(x) = 0.0695$
    • $3$ Matches: $x = 3 \implies P(x) = 0.0046$
  • Verification (Drill 10):
    • Each probability satisfies $0 \le P(x) \le 1$
    • $\sum P(x) = 0.5787 + 0.3472 + 0.0695 + 0.0046 = 1.0000$
    • The distribution is valid.
  • Graphical Representation:
    • Horizontal axis: Represents the possible profit values $x$ ($-1$, $1$, $2$, $3$).
    • Vertical axis: Represents the probability height $P(x)$.
    • Bar heights correspond to $P(-1) \approx 0.58$, $P(1) \approx 0.35$, $P(2) \approx 0.07$, and $P(3) \approx 0.005$

Mean of a Discrete Random Variable

• Definition & Concept:
  • The mean of a discrete random variable $X$ (denoted as $\mu_X$) represents the expected long-term average value of $X$ over a theoretically infinite number of repeated trials.
• Formula:     $\mu_X = \sum [x \times P(x)]$
• Calculation Procedure:
  1. Take each possible value $x$ of the random variable.
  2. Multiply each value $x$ by its corresponding probability $P(x)$.
  3. Sum all resulting products together.
• Chuck-a-Luck Expected Mean Calculation:
  • Using the probability distribution values:         $\mu_X = (-1 \times 0.5787) + (1 \times 0.3472) + (2 \times 0.0695) + (3 \times 0.0046)$
  • Step-by-step expansion:         $\mu_X = -0.5787 + 0.3472 + 0.1390 + 0.0138$$\mu_X = -0.0787$
• Standard Interpretation Rules:
  • Interpretation Phrase Format: "In the long term (or long run), you would lose $\$0.0787$ per game on average."
  • Critical Requirement: Any interpretation of $\mu_X$ MUST explicitly state the phrase "on average" to be mathematically correct.