[ Home | GAUSS | GAUSS Engine | GAUSS Apps | 3rd Party Apps | Keywords Index ]

Transforming Independent Variables

setCVIndEqs
Next ->: The Estimation

The keyword function setIndVars installs the selected independent variables in the FANPAC global _fan_IndVars. Once this has been accomplished you can directly transform the variables in that matrix. Thus

    library fanpac;
    session amzn 'inCV model';
    setDataSet stocks;
    setSeries AMZN;

    setIndVars AMZNvol;

    _fan_IndVars = ln(_fan_IndVars);


    estimate run2 garch(1,1);

It is important that any such transformations take place after all calls to setSeries, setIndVars, and computeLogReturns or computePercentReturns. And transformations must not change the length of the data matrix stored in _fan_IndVars.

Generally, it would be safer and better to make any required transformations to the data in a separate command file in which the new dataset for analysis is created.

• Basic Use
  • Initializing the Session
  • Defining the Dataset
  • The Dependent Variable
    • Transforming the Dependent Variable
    • Direct Transformation
  • The Independent Variables
    • setIndVars
    • setIndEqs
    • setCVIndEqs
    • Transforming Independent Variables
  • The Estimation
  • Displaying Results
    • Conditional Variances, Standard Deviations, Residuals
    • Forecasts
• Advanced Use
• Session File
• Modifying NLP globals
• Start Values
  • Saving and Retrieving Estimates for Start Values
    • Retrieval of Start Values
    • Direct Retrieval of Estimates
• Working with NLP
• Appendix A
  • Session Entries
  • Run Entries
• About this document

R. Schoenberg
1999-03-29