Logistic regression is similar to his example on fitting a line to an exponential trend. The differences is that in logistic regression, the equation would be

Logistic regression is similar to his example on fitting a line to an exponential trend. The differences is that in logistic regression, the equation would be ln(p)/ln(1 - p) = α + Σßᵢxᵢ where p is in the interval [0, 1] and E is . Note: in logistic regression, the notation is generally p (as in probability of event), not y.