TOPICS
Search

Probit Function


The probit function is the quantile function of the standard normal distribution. For a probability 0<p<1, it is given by

 probit(p)=Phi^(-1)(p)=sqrt(2)erf^(-1)(2p-1),

where Phi(z) is the cumulative distribution function of the standard normal distribution and erf^(-1)(x) is the inverse erf function. In the convention of Bliss (1934), the same quantile is shifted by 5, so that a probability of 1/2 corresponds to 5.

The probit function serves as a link for binary responses in a generalized linear model. With this link, the response probability is Phi(eta), where eta is the linear predictor.

The probit function will be implemented in a future version of the Wolfram Language as Probit[p].


See also

Generalized Linear Model, Inverse Erf, Logit Function, Quantile Function, Standard Normal Distribution

Explore with Wolfram|Alpha

References

Bliss, C. I. "The Method of Probits--A Correction." Science 79, 409-410, 1934. https://doi.org/10.1126/science.79.2053.409.

Cite this as:

Weisstein, Eric W. "Probit Function." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/ProbitFunction.html

Subject classifications