Which modeling approach maps input factors to a default probability via a logistic function?

• A. Linear Probability Model
• B. Logit Model
• C. Probit Model
• D. Neural Network

Answer: (B)

A logistic regression (the logit model) uses a linear combination of input factors that is transformed by the logistic function to produce a probability. Specifically, it computes p(default) = 1 / (1 + exp(-Xβ)). This keeps predicted probabilities between 0 and 1 and allows interpretation in terms of log-odds: each predictor changes the log-odds of default by its coefficient. This approach is preferred over a linear probability model, which can yield probabilities outside the 0–1 range and misrepresent variance. While the probit model also maps to a probability, it uses a different link function (the standard normal CDF) rather than the logistic function. Neural networks can produce probabilities too, but they aren’t defined by a single logistic link in the same way.