Which is true about logistic regression as compared to ordinary least squares regression?

Logistic regression is specifically designed for situations where the dependent variable is binary, meaning it takes on two possible outcomes, such as success/failure or yes/no. This method models the probability that a given data point belongs to a certain category by using the logistic function, which outputs values between 0 and 1. This characteristic makes logistic regression particularly suitable for binary outcomes, as it directly addresses the challenge of estimating the probability of an event occurring.

In contrast, ordinary least squares (OLS) regression is used for continuous dependent variables, and it operates under different assumptions, such as normally distributed residuals. OLS is not appropriate for binary outcomes since it can produce predicted values that fall outside the logical range of probabilities (0 to 1), which does not make sense in the context of binary classification.

Therefore, recognizing the specific application of logistic regression for binary dependent variables reinforces its importance in statistical modeling for classification tasks, which is why this choice is correct.