LikPak provides a set of GAUSS likelihood procedures that are commonly used in econometrics, and show, by example, how a model can be parameterized using these likelihoods. Thus LikPak is the perfect companion to an optimization package, such as MaxLik or CML. LikPak is designed to be used as a template - that is, you select the example that is relevant to your problem, and use that example as a starting point. There are over 50 likelihood functions in the LikPak package, corresponding to the set of likelihoods currently used in economics. Each example is backed up with documentation describing typical parameterizations. In addition, since there are a number of different optimization tools available, these examples are repeated for each optimization tool. Other features include both truncation and censoring for statistical functions. For least squares problems, LikPak provides the NLS command for single and multiple non-linear equation systems. For languages requiring structures, LikPak provides a set of (optional) PV and DS commands that simplify the use of the PV and DS structures, as well as facilitating passing options. And additional utilities are provided for data generation, filtering, Gibbs sampling, and constraining parameters. The source code is written in GAUSS, and will run on any version of GAUSS or the GAUSS Engine.
Comments