Business Analytics 101: Probability and Distributions Random Variables Discrete vs Continuous Expected Value Variance

What This Is

Random variables are a fundamental concept in business analytics, allowing us to model and analyze uncertain events or outcomes. In business, random variables are used to forecast sales, segment customers, detect fraud, and make informed decisions under uncertainty. For example, a retail company might use a random variable to model the daily demand for a new product, taking into account factors like seasonality, competition, and marketing campaigns.

Key Formulas & Metrics

• Discrete Random Variable (X): A random variable that can take on a countable number of distinct values. For example, the number of customers visiting a store on a given day.
• Probability Mass Function (PMF): P(X = x) = probability that X takes on the value x.
• Expected Value (E(X)): E(X) = ∑xP(X = x) – average value of the random variable.

• Variance (Var(X)): Var(X) = E(X²) – (E(X))² – measures the spread of the random variable.

• Continuous Random Variable (X): A random variable that can take on any value within a given interval or range.

• Probability Density Function (PDF): f(x) = probability density of the random variable at x.
• Expected Value (E(X)): E(X) = ∫xf(x)dx – average value of the random variable.

• Variance (Var(X)): Var(X) = E(X²) – (E(X))² – measures the spread of the random variable.

• Expected Value of a Function (g(X)): E[g(X)] = ∑g(x)P(X = x) (discrete) or ∫g(x)f(x)dx (continuous) – average value of the function g(X).

• Covariance (Cov(X, Y)): Cov(X, Y) = E(XY) – E(X)E(Y) – measures the linear relationship between X and Y.

• Correlation Coefficient (ρ(X, Y)): ρ(X, Y) = Cov(X, Y) / (σXσY) – measures the strength and direction of the linear relationship between X and Y.

Step-by-Step Procedure

1. Define the Random Variable: Identify the uncertain event or outcome you want to model.
2. Determine the Type of Random Variable: Is it discrete or continuous?
3. Specify the Probability Distribution: Choose a probability distribution that fits the data (e.g., normal, Poisson, binomial).
4. Estimate the Parameters: Use data to estimate the parameters of the probability distribution (e.g., mean, variance).
5. Compute the Expected Value and Variance: Use the probability distribution to compute the expected value and variance of the random variable.

Common Mistakes

• Mistake: Confusing correlation with causation.
• Correction: Correlation does not imply causation. Use techniques like regression analysis to establish causality.
• Mistake: Misinterpreting p-values.
• Correction: A small p-value indicates that the observed data is unlikely under the null hypothesis, but it does not prove the null hypothesis is false.
• Mistake: Using the wrong error metric for a business problem.
• Correction: Choose an error metric that aligns with the business objective (e.g., mean absolute error for forecasting).

Software / Tool Tips

• Python with pandas/scikit-learn: Use the numpy library to compute expected values and variances, and the scipy.stats library to specify probability distributions.
• R: Use the stats package to compute expected values and variances, and the distributions package to specify probability distributions.
• Excel: Use the AVERAGE and VAR functions to compute expected values and variances.

Quick Practice Problem

Problem: Compute the expected value of a random variable X that follows a normal distribution with mean 5 and variance 2.

Answer: E(X) = 5.

Explanation: The expected value of a normal distribution is equal to the mean.

Last-Minute Cram Sheet

1. A discrete random variable can take on a countable number of distinct values.
2. The expected value of a random variable is a measure of its central tendency.
3. The variance of a random variable measures its spread.
4. The covariance between two random variables measures their linear relationship.
5. The correlation coefficient measures the strength and direction of the linear relationship between two random variables.
6. A small p-value indicates that the observed data is unlikely under the null hypothesis.
7. Correlation does not imply causation.
8. The expected value of a function of a random variable is equal to the integral of the function times the probability density function.
9. The variance of a function of a random variable is equal to the integral of the square of the function times the probability density function minus the square of the expected value of the function.
10. ⚠️ p-value is NOT the probability that H₀ is true – it’s the probability of observing the data (or more extreme) if H₀ is true.