Week 6 - Comprehensive Study Notes on Discrete Random Variables

The Role and Impact of Random Variables

• Conceptual Foundations: Why Random Variables Matter
  • Quantifying Uncertainty: Random variables serve to assign numerical values to random outcomes, which transforms abstract uncertainty into a measurable and manageable format.
  • Computing Probabilities: By utilizing probability distributions, it becomes possible to determine the likelihood of specific events, enabling precise calculations rather than estimations.
  • Predictions and Decision-Making: These tools allow for the forecasting of outcomes and the rigorous assessment of risk, which directly supports and enhances high-level decision-making processes.
  • Modeling Real-World Phenomena: They provide a mathematical framework to represent the complexities of reality through structured distributions, forming the bedrock of statistical inference and modern data science.
  • Foundations for Expectation: They define the "expected value," which is crucial for understanding the long-term behavior of a system or process.
  • Measuring Variability and Risk: Through metrics like variance and standard deviation, random variables quantify the volatility and spread of potential outcomes.

Learning Objectives for Discrete Probability Distributions

• Relational Understanding: Students must understand the relationship between specific outcomes and their associated probabilities within a distribution.
• Descriptive Methods: Proficiency is required in describing discrete probability distributions using three distinct formats:
  • Tables: Listing individual outcomes alongside their probabilities.
  • Graphs: Visual representations of probability mass functions.
  • Formulas: Mathematical expressions (PMFs) that define the distribution.
• Analytical Computation: Students must know how to compute and interpret the Expected Value and Standard Deviation of these distributions.

Fundamental Definitions and Classifications

• Random Variable ($X$): A numerical description of the outcome resulting from an experiment.
• Discrete Random Variable:
  • Defined by its ability to assume either a finite number of values or an infinite sequence of countable values.
  • Examples:
    • Hotel Reservations: Making $100$ reservations; $X = \text{number of guests who show up}$. Possible outcomes: $0, 1, 2, \bar{\text{...}}, 98, 99, 100$.
    • Restaurant Operations: Running a restaurant for one day; $X = \text{number of customers}$. Possible outcomes: $0, 1, 2, 3, \bar{\text{...}}$.
    • Room Sales: Choosing a hotel room; $X = 1$ if it has a sea view, $X = 0$ if it does not.
• Continuous Random Variable:
  • Defined by its ability to assume any numerical value within an interval or a collection of intervals.
  • Example:
    • Hotel Reception: $X = \text{time between two consecutive customer arrivals, in minutes}$. Possible outcomes: $x \text{ } \boldsymbol{\beta} \text{ } 0$.

The Discrete Probability Distribution

• Definition: This distribution describes how probabilities are assigned across the possible values of a discrete random variable.
• Representations:
  • Probability Mass Function (pmf): Denoted as $f(x)$, it provides the probability for each specific value of the random variable.
  • Alternative Notation: $P(X = x)$, which explicitly states the probability that the variable $X$ takes the value $x$.
• Mathematical Requirements: For a discrete random variable, the following conditions must always be satisfied:
  1. Individual probabilities must be non-negative: $f(x) \boldsymbol{\beta} 0$ and $P(X = x) \boldsymbol{\beta} 0$.
  2. The sum of all probabilities in the sample space must equal exactly one: $\text{SUM}(f(x)) = 1$ or $\text{SUM}(P(X = x)) = 1$.

Case Study: Uniform Distribution (Rolling a Single Die)

• Experiment: Rolling one standard six-sided die.
• Random Variable ($X$): The number of dots on the top face.
• Possible Values ($x$): $1, 2, 3, 4, 5, 6$.
• Probability Representation:
  • Table: Every outcome ($1$ through $6$) has an identical probability of $P(X = x) = \frac{1}{6}$.
  • Graph: A bar chart where each outcome from $1$ to $6$ reaches the height of approximately $0.167$.
  • Formula: $f(x) = \frac{1}{n}$, where $n$ is the number of possible outcomes. For a die, $f(x) = \frac{1}{6}$.

Case Study: Distribution of the Sum of Two Dice

• Experiment: Rolling two standard dice simultaneously.
• Random Variable ($X$): The sum of the number of dots on the top faces of both dice.
• Possible Values: Values range from $2$ (rolling two 1s) to $12$ (rolling two 6s), specifically: $2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12$.

Measures of Central Tendency: Expected Value

• Theoretical Concept: The expected value is the theoretical long-term average of a random variable. It does not necessarily represent an outcome that will occur in a single trial.
• Formula:
  • $E[X] = \text{SUM}(x_i \times P(X = x_i))$
• Excel Implementation: Use the function SUMPRODUCT(values; probabilities).
• Practical Use: This is a vital tool for decision-making under uncertainty, as it only requires knowledge of probabilities and outcomes, not historical data.
• Mathematical Properties:
  • Linearity: $E[aX + b] = a \times E[X] + b$
  • Additivity: $E[X + Y] = E[X] + E[Y]$

Real-World Example: Stock Decisions Based on Expected Value

• Scenario: A stock trades at $250$. Overnight results will determine tomorrow's price.
  • Good News (60% probability): Price rises to $255$ (Profit = $+5$).
  • Bad News (40% probability): Price falls to $240$ (Profit = $-10$).
• Expected Price Calculation:
  • $255 \times 0.60 + 240 \times 0.40 = 153 + 96 = 249$
• Expected Profit Calculation:
  • $5 \times 0.60 + (-10) \times 0.40 = 3 - 4 = -1$
• Decision: Based on an expected profit of $-1$, you should not buy the stock.

The Gambler’s Fallacy: Roulette Example

• Observation: The last $6$ trials at a roulette table resulted in: $2, 32, 14, 27, 30, 21$. Out of these, $2$ outcomes were "Black."
• Question: What is the most likely number of "Black" outcomes considering a total of $8$ spins (the past $6$ results plus $2$ future spins)?
• Analysis of Future Spins: For the next $2$ spins, there are four possible combinations, each with a probability of $0.25$ ($0.5 \times 0.5$):
  • Red, Red ($0$ Blacks): $25$%
  • Red, Black / Black, Red ($1$ Black): $50$%
  • Black, Black ($2$ Blacks): $25$%
• Conclusion: The most likely outcome for the future spins is $1$ "Black." Added to the $2$ that already occurred, the most likely total is $3$.

Measures of Variability: Variance and Standard Deviation

• Variance Formula ($Var(X)$):
  • $Var(X) = \text{SUM}((x_i - E[X])^2 \times P(X = x_i))$
  • Excel Implementation: SUMPRODUCT((values - expected_value)^2; probabilities).
• Standard Deviation Formula (̓(X)):
  • ̓(X) = \text{SQRT}(Var(X))
• Use and Interpretation:
  • Measures the dispersion of outcomes around the expected value.
  • Standard Deviation is expressed in the same units as the random variable ($X$), whereas Variance is in squared units.
• Additivity Property:
  • $Var(X + Y) = Var(X) + Var(Y)$

Summary of Common Discrete Probability Distributions