Multinomial Probit - General Specification

General Specification

It is assumed that we have a series of observations Yi, for i=1...n, of the outcomes of multi-way choices from a categorical distribution of size m (there are m possible choices). Along with each observation Yi is a set of observed explanatory variables x1,i, ..., xk,i (aka independent variables, predictor variables, features, etc.). Some examples:

  • The observed outcomes might be "has disease A, has disease B, has disease C, has none of the diseases" for a set of rare diseases with similar symptoms, and the explanatory variables might be characteristics of the patients thought to be pertinent (sex, race, age, blood pressure, body-mass index, presence or absence of various symptoms, etc.).
  • The observed outcomes are the votes of people for a given party or candidate in a multi-way election, and the explanatory variables are the demographic characteristics of each person (e.g. sex, race, age, income, etc.).

The multinomial probit model seeks to construct a statistical model that can be used to predict the likely outcome of another unobserved multi-way trial given the associated explanatory variables. In the process, the model attempts to explain the relative effect of differing explanatory variables on the different outcomes.

Formally, the outcomes Yi are described as being categorically-distributed data, where each outcome h is determined by an unobserved probability pi,h that is specific to the outcome at hand, but related to the explanatory variables. That is:

or equivalently

where denotes the expected value of Yi.


Read more about this topic:  Multinomial Probit

Famous quotes containing the word general:

    In effect, to follow, not to force the public inclination; to give a direction, a form, a technical dress, and a specific sanction, to the general sense of the community, is the true end of legislature.
    Edmund Burke (1729–1797)