To predict the probability of a binary outcome, which model is most appropriate?

• A. Linear Regression
• B. K-Means
• C. Logistic Regression
• D. Decision Tree

Answer: (C)

Modeling the probability of a binary outcome requires a method that naturally links predictors to a probability between 0 and 1. Logistic regression does this by applying a logistic function to a linear combination of the predictors, so the predicted probability P(Y=1|X) always falls in (0,1). The coefficients are estimated via maximum likelihood, and they express how the log-odds of the event change with each predictor, which gives clear, interpretable insight. This approach avoids the issue with linear regression, which can produce predictions outside the 0–1 range and relies on assumptions that don’t fit binary data. K-means is an unsupervised clustering method, not designed to predict binary outcomes or probabilities. A decision tree can estimate probabilities, but logistic regression provides a principled, widely used probabilistic model with straightforward interpretation and statistical inference.