BRFPREDICT procedure

Makes predictions using a random regression forest (R.W. Payne).

Options

`PRINT` = string token	Controls printed output (`predictions`); default `pred`
`PREDICTIONS` = variate	Saves the prediction for the observations
`SAVE` = pointer	Save structure from `BRFOREST` providing information about the random forest

Parameters

`X` = variates or factors	Explanatory variables
`VALUES` = scalars, variates or texts	Values to use for the explanatory variables; if these are unset for any variable, its existing values are used

Description

BRFPREDICT makes predictions using a regression forest, as constructed by the BRFOREST procedure. The SAVE parameter can be set to a pointer, saved using the SAVE option of BRFOREST, containing the necessary information about the forest. Alternatively, if you do not set SAVE, BRFPREDICT will use the forest most recently constructed by BRFOREST.

The x-values for the predictions can be specified in the variates listed by the X parameter. These must have identical names (and levels) to those used originally to construct the tree. You can use the VALUES parameter to supply new values, if those stored in any of the variates or factors are unsuitable.

By default, BRFPREDICT prints the predictions, but you can set option PRINT=* to suppress this. The PREDICTIONS option allows you to save the predictions.

Options: PRINT, PREDICTIONS, SAVE.
Parameters: X, VALUES.

Method

BRFPREDICT takes the mean of predictions from the individual trees, made using BIDENTIFY.

Action with `RESTRICT`

Restrictions are ignored.

Example

CAPTION    'BRFPREDICT example'; STYLE=meta
SPLOAD     [PRINT=*] '%gendir%/data/water.gsh'
BRFOREST   [PRINT=outofbagerror,youtofbagestimates,importance;\
           Y=Water; NTREES=8; NXTRY=3; NUNITSTRY=10; SEED=185090]\
           Employ,Opdays,Product,Temp
BRFPREDICT [PREDICTION=prediction] Opdays,Product,Temp;\
           VALUES=!(20,23,20),!(11.96,13.53,12.66),!(71.90,63.90,54.50)

Updated on February 6, 2023

Was this article helpful?

Yes No