The RELIABILITY Procedure |

LOGSCALE Statement |

- LOGSCALE effect-list < /options > ;

You use the LOGSCALE statement to model the logarithm of the distribution scale parameter as a function of explanatory variables. A MODEL statement must also be present to specify the model for the distribution location parameter. *effect-list* is a list of variables in the input data set representing the values of the independent variables in the model for each observation, and combinations of variables representing interaction terms. It can contain any variables or combination of variables in the input data set. It can contain the same variables as the MODEL statement, or it can contain different variables. The variables in the *effect-list* can be any combination of indicator variables named in a CLASS statement and continuous variables. The coefficients of the explanatory variables are estimated by maximum likelihood.

The following *options* are available for the LOGSCALE statement.

Option |
Option Description |
---|---|

INITIAL= |
specifies initial values for log-scale regression parameters other than the intercept term |

INTERCEPT= |
specifies initial or fixed value of the intercept parameter, depending on whether INTINIT is present |

Copyright © SAS Institute, Inc. All Rights Reserved.