count |

Estimates models where the dependent variable is a nonnegative integer count.

Syntax

count(options) y x1 [x2 x3...]

count(options) specification

Follow the count keyword by the name of the dependent variable and a list of regressors.

Options

d=arg (default=“p”) | Likelihood specification: Poisson likelihood (“p”), normal quasi-likelihood (“n”), exponential likelihood (“e”), negative binomial likelihood or quasi-likelihood (“b”). |

v=positive_num (default=1) | Specify fixed value for QML parameter in normal and negative binomial quasi-likelihoods. |

optmethod = arg | Optimization method: “bfgs” (BFGS); “newton” (Newton-Raphson), “opg” or “bhhh” (OPG or BHHH), “legacy” (EViews legacy). Newton-Raphson is the default method. |

optstep = arg | Step method: “marquardt” (Marquardt); “dogleg” (Dogleg); “linesearch” (Line search). Marquardt is the default method. |

cov=arg | Covariance method: “ordinary” (default method based on inverse of the estimated information matrix), “huber” or “white” (Huber-White sandwich methods)., “glm” (GLM method).. |

covinfo = arg | Information matrix method: “opg” (OPG); “hessian” (observed Hessian). (Applicable when non-legacy “optmethod=”.) |

h | Huber-White quasi-maximum likelihood (QML) standard errors and covariances. (Legacy option Applicable when “optmethod=legacy”). |

g | GLM standard errors and covariances. (Legacy option Applicable when “optmethod=legacy”). |

m=integer | Set maximum number of iterations. |

c=scalar | Set convergence criterion. The criterion is based upon the maximum of the percentage changes in the scaled coefficients. The criterion will be set to the nearest value between 1e-24 and 0.2. |

s | Use the current coefficient values in “C” as starting values (see also
param). |

s=number | Specify a number between zero and one to determine starting values as a fraction of the EViews default values (out of range values are set to “s=1”). |

prompt | Force the dialog to appear from within a program. |

p | Print the result. |

Examples

The command:

count(d=n,v=2,cov=glm) y c x1 x2

estimates a normal QML count model of Y on a constant, X1, and X2, with fixed variance parameter 2, and GLM standard errors.

count arrest c job police

makeresid(g) res_g

estimates a Poisson count model of ARREST on a constant, JOB, and POLICE, and stores the generalized residuals in the series RES_G.

count(d=p) y c x1

fit yhat

estimates a Poisson count model of Y on a constant and X1, and saves the fitted values (conditional mean) in the series YHAT.

count(d=p, h) y c x1

estimates the same model with QML standard errors and covariances.

Cross-references

See
“Count Models” for additional discussion.

See
Equation::count for the equivalent equation object command.