Regression Action Set

This section contains Lua code that creates the same analysisData table that is shown in the CASL example. It highlights sections of the example where the Lua syntax differs significantly from the CASL syntax. See the CASL version of this example for discussions of the data, syntax features, and the output from running these examples. For more information about coding in Lua, see Getting Started with SAS Viya for Lua and SAS Viya: System Programming Guide.

Because this example uses simulated data that are produced from DATA step code, you can submit the runCode action on your Lua client as follows to create the simulated data in your CAS session:

s:runCode{single='YES',code=[[
   data analysisData;
      call streamInit(1);
      do id=1 to 100;
         x1 = rand('UNIFORM');
         x2 = rand('NORMAL');
         c1 = 1 + mod(id,3);
         c2 = rand('BERNOULLI',0.6);
         y  = 1 + x1 + c1 + 0.01*rand('NORMAL');
         output;
      end;
   run;
   ]]
}

The following code loads the regression action set and uses the glm action to fit a linear regression model to these data by using forward selection and to store the model:

s:loadactionset{actionset="regression"}
mfit=s:glm{
   table='analysisData',
   class={'c1','c2'},
   model={depvar='y',
          effects={'c1', 'c2', 'x1', 'x2'}},
   store={name='myModel', replace=true},
   selection={method='forward'},
   output={casOut={name='out1', replace=true},
           copyVars='ALL',
           pred='pred', resid='resid', rstudent='rstudent'} }

The store parameter stores the model in a special data table named myModel. The output parameter scores the observations in the analysisData data table and writes to the out1 data table the predicted values, residuals, studentized residuals with the current observation deleted, and all the variables from the original data table.

The following command displays information about the out1 table:

print(mfit.OutputCasTables)

The following commands use the glmScore action in the regression action set to score the original data set, using the model that was previously fit:

mscore=s:glmScore{
   table='analysisData',
   restore='myModel',
   casOut={name='out2', replace=true},
   fitData='true',
   copyVars='ALL',
   pred='pred',
   resid='resid',
   rstudent='rstudent' }

The table parameter names the table to be scored, the restore parameter names the stored model to be used for scoring, the casOut parameter names the output data table, the copyVars='ALL' parameter requests that all variables in the analysisData data table be copied to the out2 output data table, and the remaining parameters name the statistics to store in the output data table.

The following command displays information about the out2 table:

print(mscore.OutputCasTables)

For details about the results of this analysis, see the CASL version of this example.

This section contains Python code that creates the same analysisData table that is shown in the CASL example. It highlights sections of the example where the Python syntax differs significantly from the CASL syntax. See the CASL version of this example for discussions of the data, syntax features, and the output from running these examples. For more information about coding in Python, see Getting Started with SAS Viya for Python and SAS Viya: System Programming Guide.

Because this example uses simulated data that are produced from DATA step code, you can submit the runCode action on your Python client as follows to create the simulated data in your CAS session:

s.runCode(single='YES',code="""
   data analysisData;
      call streamInit(1);
      do id=1 to 100;
         x1 = rand('UNIFORM');
         x2 = rand('NORMAL');
         c1 = 1 + mod(id,3);
         c2 = rand('BERNOULLI',0.6);
         y  = 1 + x1 + c1 + 0.01*rand('NORMAL');
         output;
      end;
   run;"""
)

The following code loads the regression action set and uses the glm action to fit a linear regression model to these data by using forward selection and to store the model:

s.loadactionset(actionset="regression")
mfit=s.glm(
   table='analysisData',
   classvars={'c1','c2'},
   model={'depvar':'y',
          'effects':['c1', 'c2', 'x1', 'x2']},
   store={'name':'myModel', 'replace':'True'},
   selection={'method':'forward'},
   output={'casOut':{'name':'out1', 'replace':'True'},
           'copyVars':'ALL',
           'pred':'pred', 'resid':'resid', 'rstudent':'rstudent'}
)

The store parameter stores the model in a special data table named myModel. The output parameter scores the observations in the analysisData data table and writes to the out1 data table the predicted values, residuals, studentized residuals with the current observation deleted, and all the variables from the original data table.

The following command displays information about the out1 table:

print(mfit.OutputCasTables)

The following code uses the glmScore action in the regression action set to score the original data set, using the model that was previously fit:

mscore=s.glmScore(
   table='analysisData',
   restore='myModel',
   casOut={'name':'out2', 'replace':'True'},
   fitData='true',
   copyVars='ALL',
   pred='pred',
   resid='resid',
   rstudent='rstudent'
)

The table parameter names the table to be scored, the restore parameter names the stored model to be used for scoring, the casOut parameter names the output data table, the copyVars='ALL' parameter requests that all variables in the analysisData data table be copied to the out2 output data table, and the remaining parameters name the statistics to store in the output data table.

The following command displays information about the out2 table:

print(mscore.OutputCasTables)

For details about the results of this analysis, see the CASL version of this example.

This section contains R code that creates the same analysisData table that is shown in the CASL example. It highlights sections of the example where the R syntax differs significantly from the CASL syntax. See the CASL version of this example for discussions of the data, syntax features, and the output from running these examples. For more information about coding in R, see Getting Started with SAS Viya for R and SAS Viya: System Programming Guide.

Because this example uses simulated data that are produced from DATA step code, you can submit the runCode action on your R client as follows to create the simulated data in your CAS session:

m <- cas.builtins.loadActionSet(s, actionset='dataStep')
m <- cas.dataStep.runCode(s,single='YES',code="
   data analysisData;
      call streamInit(1);
      do id=1 to 100;
         x1 = rand('UNIFORM');
         x2 = rand('NORMAL');
         c1 = 1 + mod(id,3);
         c2 = rand('BERNOULLI',0.6);
         y  = 1 + x1 + c1 + 0.01*rand('NORMAL');
         output;
      end;
   run;"
)

The following code loads the regression action set and uses the glm action to fit a linear regression model to these data by using forward selection and to store the model:

m <- cas.builtins.loadActionSet(s, actionset='regression')
mfit <- cas.regression.glm(s,
   table='analysisData',
   class=list('c1','c2'),
   model=list(depvar='y',
              effects=list('c1', 'c2', 'x1', 'x2')),
   store=list(name='myModel', replace=TRUE),
   selection=list(method='forward'),
   output=list(casOut=list(name='out1', replace=TRUE),
               copyVars='ALL',
               pred='pred', resid='resid', rstudent='rstudent')
)

The store parameter stores the model in a special data table named myModel. The output parameter scores the observations in the analysisData data table and writes to the out1 data table the predicted values, residuals, studentized residuals with the current observation deleted, and all the variables from the original data table.

The following command displays information about the out1 table:

print(mfit$OutputCasTables)

The following code uses the glmScore action in the regression action set to score the original data set, using the model that was previously fit:

mscore <- cas.regression.glmScore(s,
   table='analysisData',
   restore='myModel',
   casOut=list(name='out2', replace=TRUE),
   fitData=TRUE,
   copyVars='ALL',
   pred='pred',
   resid='resid',
   rstudent='rstudent'
)

The table parameter names the table to be scored, the restore parameter names the stored model to be used for scoring, the casOut parameter names the output data table, the copyVars='ALL' parameter requests that all variables in the analysisData data table be copied to the out2 output data table, and the remaining parameters name the statistics to store in the output data table.

The following command displays information about the out2 table:

print(mscore$OutputCasTables)

For details about the results of this analysis, see the CASL version of this example.