Third-Party Applications

Markov

The following product is developed by a third party company, for use with GAUSS. Technical support is provided directly through the developer.

Markov 2.5

A Statistical Environment for GAUSS 3.2
J. Scott Long

Markov is a statistical environment that makes it easy to do simple things without restricting your use of the full power of GAUSS. Experienced GAUSS programmers can work more efficiently. New users will find that Markov makes GAUSS easier and more fun to learn. A user writes: "Markov makes life a lot easier for the GAUSS user, even for people who have learned how to do things the harder way in GAUSS." Markov is ideal for classroom use where you want students to be able to get output quickly and easily, but also want them to have access to a matrix language

How Markov Works
Markov has a simple command structure. For example, to run a multiple regression with collinearity diagnostics you would use the commands:

SET DSN MYDATA;
SET LHS YVAR;
SET RHS XVAR1 XVAR2;
OPT COLLIN ON;
GO REG;

After running an analysis the results are printed and returned to global variables that can be used in your own GAUSS programs or can be further analyzed by Markov. For example, to test the hypothesis that the coefficients for XVAR1 and XVAR2 are both zero, you would simply enter the commands

TEST DELETE XVAR1 XVAR2

Complicated graphs can be computed just as easily, with impressive results:


Features
Markov includes extensive on-line documentation, a 200 page manual that includes complete technical documentation, and source code so that you can customize Markov.

Statistical procedures include descriptive statistics, cross-tabulation, log-linear models, multinomial logit, probit, Poisson regression, ordered logit and probit, tobit, switching regressions, simultaneous equation models, and regression analysis with collinearity diagnostics, residual analysis and powerful statistical tests such as White's information matrix test.

Statistical graphics include box and whisker plots, scatterplot matrices, quantile-quantile plots, and many more. Each plot is specified with a simple command language.

A Shell for Monte Carlo simulation is included. This allows GAUSS programmers to do complex simulations very simply. You program the procedure you want to simulate, and Markov keeps track of the results of the simulation and allows simulations to be suspended and resumed later.

Full data management capabilities including sorting, merging and updating. Variables in memory can be saved to GAUSS data files as simply as entering the command:

WRITE X Y Z TO NEWFILE

Variables from disk files can be brought into memory simply:

READ WEIGHT LENGTH FROM AUTO

Enhancements to GAUSS's DATALOOP procedure make it easy to construct the types of variables most commonly used instatistical analysis.

What's New in Version 2.5?

The biggest change is that Markov has been converted from GAUSS 3.1 for DOS to GAUSS 3.2 for Windows. While the program was written using GAUSS for Windows Version 3.2, it should also work on other platforms. The change from the DOS based GAUSS to the Window's version required many changes to the way information was written to the screen, especially for procedures like MONTE that displayed a lot of interactive information. In addition to change required for GAUSS for Windows, other enhancements have been added. These include:

  • The tobit and endogenous switching regression models have been added.
  • The output is more compact and hopefully more readable.
  • The output for single equation models, such as logit, probit and Poisson regression, is more uniform. Additional information is now returned to global matrices.
  • All format statements have been removed. This means, that when you are finished running a Markov command, your format has not been changed.
  • Three commands aid in the interpretation of models for categorical and limited dependent variables. These are PREDICT which computes the predicted value of the outcome; PARTIAL which computes the partial derivative of the outcome; and DISCRETE which computes discrete changes in the outcome for given changes in the independent variables.