AP Statistics Unit 4: Binomial and Geometric Distributions Review

Overview of the Binomial Distribution

• Definition: The binomial distribution is a probability distribution for a specific type of discrete random variable, where outcomes are countable numerical values.
• Success-Failure Model: It represents a random variable $X$, which equals the number of successes in $n$ repeated trials of a random process where only two outcomes are possible for each trial (success or failure).
• Definition of Success: In this context, "success" is simply the event or outcome the researcher is looking for. It does not necessarily represent a positive outcome; it could be something undesirable, as long as it is the focus of the count.
• Components for Naming: The prefix "bi" in binomial refers to the two essential pieces of information required to define the distribution:     * $P$: The probability of success.     * $n$: A set number of trials or opportunities.

Rules and Requirements for Binomial Distributions

• Clearly Defined Success: The success must be explicitly defined, and its probability must be given.
• Constant Probability: The probability of success, denoted as $P$, must remain the same for every single trial or opportunity throughout the process.
• The Complement: Once the probability of success $P$ is known, the probability of failure is automatically known as its complement, $1 - P$.
• Independence: Each trial or opportunity must be independent of the others. The outcome of one trial (success or failure) cannot affect the outcome of any subsequent trials.
• Fixed Number of Trials: There must be a specific, given number of trials or opportunities, denoted as $n$.
• Distribution Requirements: Like all probability distributions, a binomial distribution must include:     * A comprehensive list of all possible outcomes for $X$.     * The probability associated with each of those outcomes.

Generic Case Study and Possible Outcomes

• Scenario: Consider a process with $n = 7$ trials and a probability of success $P = 0.22$ ($22\%$) for any given trial.
• List of Possible Outcomes for X: Because there are only 7 trials, the number of successes $X$ can only be whole numbers ranging from 0 to 7:     * $X \in {0, 1, 2, 3, 4, 5, 6, 7}$.
• Inherent Constraints: It is impossible to have more successes than trials (e.g., 8, 9, or 10 successes) or a non-integer number of successes (e.g., $2.5$ successes).

Binomial Distribution Application: David’s Frequent Flyer Upgrades

• Context: David is a member of a frequent flyers club. On any individual flight, he has a $22\%$ chance of having his seat upgraded to first class.
• Trial Parameters: The scenario looks at his next $n = 7$ flights.
• Rule Verification:     * Success: Getting a seat upgrade is the success ($P = 0.22$).     * Failure: Not getting an upgrade is the failure ($1 - 0.22 = 0.78$).     * Independence: It is assumed that the chance of an upgrade on one flight is independent of all others.     * Trials: There is a fixed number of flights ($n = 7$).
• Calculating Specific Probabilities (Example: Exactly 2 Upgrades):     * One Specific Scenario: If David gets upgrades on his first and fifth flights and no upgrades on others, the probability is represented by $0.22 \times 0.22 \times 0.78 \times 0.78 \times 0.78 \times 0.78 \times 0.78$.     * Simplified Form: $(0.22)^2 \times (0.78)^5$.     * The Combination Problem: Since the two upgrades can happen in many different sequences (not just first and fifth), we must account for all ways to choose 2 successes out of 7 trials.     * Combination Formula: The number of ways is calculated using "7 choose 2," written as $\binom{7}{2}$.

Methods for Calculating the Binomial Coefficient (n choose x)

• Option 1: The Mathematic Formula:     * ${n \choose x} = \frac{n!}{x!(n-x)!}$     * For David's example ($7 \text{ choose } 2$): $\frac{7!}{2!(7-2)!} = \frac{7!}{2! \times 5!} = 21\, \text{ways}$.
• Option 2: Calculator (NCR Function):     1. Type the value for $n$ (7).     2. Navigate to MATH -> PRB (Probability).     3. Choose option 3: nCr.     4. Type the value for $x$ (2) and hit enter. Result: 21.
• Option 3: Comprehensive Calculator Shortcuts (PDF/CDF): This is often faster as it performs the entire probability calculation at once.

Binomial Formulas and Calculator Functions

• The Binomial Probability Formula:     * $P(X=x) = \binom{n}{x} \times P^x \times (1-P)^{n-x}$     * The formula consists of three parts: (1) The number of ways to have $x$ successes in $n$ trials, (2) the probability of those successes, and (3) the probability of the accompanying failures.
• Binomial PDF (Probability Density Function):     * Usage: Used to find the probability of a single, exact value of $x$.     * Calculator Navigation: 2nd -> VARS (DISTR) -> binompdf.     * Inputs: trials ($n$), $P$, $X$ value.     * Example Result: For $n=7$, $P=0.22$, and $X=2$, the result is $0.293$.
• Binomial CDF (Cumulative Distribution Function):     * Usage: Used to find the probability of getting $x$ successes or fewer ($X \le x$). It accumulates the probability of the chosen value and all outcomes below it.     * Calculator Navigation: 2nd -> VARS (DISTR) -> binomcdf.     * Example: At most 4 upgrades: Using binomcdf with $n=7$, $P=0.22$, and $X=4$ calculates $P(X=4) + P(X=3) + P(X=2) + P(X=1) + P(X=0)$.
• Example: Four or more upgrades ($X \ge 4$):     * Approach 1: Sum the individual probabilities: $P(4) + P(5) + P(6) + P(7)$.     * Approach 2 (Complement Path): The total probability is 1. The opposite of "4 or more" is "3 or less."     * Calculation: $1 - \text{binomcdf}(7, 0.22, 3)$.     * Steps: binomcdf for $X=3$ gives $0.954$. Then, $1 - 0.954 = 0.046$ or $4.6\%$.

Mean and Standard Deviation of Binomial Distributions

• Mean ($\mu_X$): Represented by the expected value in the long run.     * Formula: $\mu_X = n \times P$     * David’s Example: $7 \times 0.22 = 1.54\, \text{upgrades}$.
• Standard Deviation ($\sigma_X$): Represents the typical variation from the mean.     * Formula: $\sigma_X = \sqrt{n \times P \times (1 - P)}$     * David’s Example: $\sqrt{7 \times 0.22 \times 0.78} = 1.096\, \text{upgrades}$.
• Interpretation: After many repetitions of the 7-flight scenario, we expect an average of 1.54 upgrades, with a typical variation of approximately 1.096 upgrades.

Overview of the Geometric Distribution

• Definition: A discrete random variable $X$ where $X$ is the number of the trial on which the first success occurs.
• Ground Rules:     * Success is clearly defined and the probability $P$ is given.     * $P$ remains constant from trial to trial.     * Trials are independent.     * No set number of trials ($n$): The experiment continues indefinitely until the first success is achieved.
• Possible Outcomes: $X \in {1, 2, 3, 4, \dots}$. While the values are countable, they are technically infinite because the first success could theoretically occur on the 100th trial or later.

Calculating Geometric Probabilities

• Scenario: David has a 22% chance ($P=0.22$) of an upgrade. Let $X$ be the number of flights until he gets his first upgrade.
• Formula for the Exactly $X$-th Trial:     * $P(X=x) = (1 - P)^{x-1} \times P$     * Example: First success on the 5th flight: This requires 4 initial failures followed by 1 success.     * Calculation: $0.78 \times 0.78 \times 0.78 \times 0.78 \times 0.22 = (0.78)^4 \times 0.22 = 0.0814$.
• Unique Property: There is only one way for the first success to happen on trial $x$ (all prior trials must be failures), so combinations are not needed.
• Probability Distribution Behavior: As $X$ increases, the probability of the first success occurring on that trial decrease, approaching but never reaching zero.

Special Probability Question: Success After the Third Trial ($X > 3$)

• Scenario: What is the probability David gets his first upgrade after the third flight (4th, 5th, 6th, etc.)?
• The "Hard Way" (Complement):     * Find the probability of success on trial 1, 2, or 3 and subtract from 1.     * $P(1) = 0.22$     * $P(2) = 0.78 \times 0.22$     * $P(3) = (0.78)^2 \times 0.22$     * Sum: $0.22 + 0.1716 + 0.1338 = 0.525$.     * Total: $1 - 0.525 = 0.475$.
• The "Easy Way" (Reasoning):     * If the first success must happen after the third trial, it simply means that the first three trials must be failures.     * Calculation: $(0.78)^3 = 0.475$.

Mean and Standard Deviation of Geometric Distributions

• Mean ($\mu_X$): The average number of trials until the first success.     * Formula: $\mu_X = \frac{1}{P}$     * David’s Example: $\frac{1}{0.22} = 4.545\, \text{flights}$.
• Standard Deviation ($\sigma_X$): The variation in the number of trials until the first success.     * Formula: $\sigma_X = \frac{\sqrt{1-P}}{P}$     * David’s Example: $\frac{\sqrt{0.78}}{0.22} = 4.014\, \text{flights}$.
• Interpretation: We expect David to wait an average of 4.545 flights for his first upgrade, with a variation around that average of 4.014 flights.

Exam Strategy and Importance

• Formula Recognition: Even if using a calculator, students must recognize the components of the formulas (e.g., $\binom{n}{x}$ and $P^x$) because multiple-choice questions on the AP Exam may ask to identify the correct formulaic expression rather than a numerical answer.
• Showing Work: On free-response questions (FRQs), it is essential to show the setup/formula used, not just the calculator output.
• Difficulty Note: Unit 4 is widely considered by students and teachers to be one of the most challenging units in AP Statistics due to the complexity of probability and random variables.