Logit model

The logit model is a particular type of binary regression model. Let $Y$ be a binary random variable, and let $X$ be a vector of random variables. The logit model assumes

$P(Y=1|X=x)={\frac {\exp(x'\beta )}{1+\exp(x'\beta )}}.$

The parameter $\beta$ is usually estimated by maximum likelihood.

---

References

Amemiya, T., 1985, Advanced Econometrics, Harvard University Press.