Group: ORIGIN*HEAT 8 0.0092 R2 < MINR2
AOV16: HOMEVAL 13 0.0091 R2 < MINR2
Class: RACE*NUMCARS 11 0.0086 R2 < MINR2
Group: RACE*NUMCARS 5 0.0085 R2 < MINR2
Class: NTITLE*ORIGIN 23 0.0085 R2 < MINR2
Class: APRTMNT*RACE 8 0.0084 R2 < MINR2
Group: NTITLE*ORIGIN 6 0.0084 R2 < MINR2
Additional effects are not listed
SS and R2 Portion for Effects Chosen for the Target
This section lists the chosen input variables from the forward stepwise regression. The table is divided into the following
five columns:
Effect lists the sequentially selected effects, which are ranked by the R2 statistic.
q
DF shows the degrees of freedom associated with each model effect.
q
R2 measures the sequential improvement in the model as input variables are selected. Multiply the R2 statistic by
100 to express it as a percentage. You can interpret the R2 statistic for the LEISURE effect as: "48.27% of the
variation in the target AMOUNT is explained by its linear relationship with LEISURE". The R2 statistic for
APPAREL indicates that this effect accounts for an additional 13.44% of the target variation.
q
SS lists the sums of squares for each model effect.
q
EMS lists
the Error Mean Square, which measures variation due to either random error or to other inputs that are
not in the model. The EMS should get smaller as important inputs are added to the model. Note that although
STATECOD has an R2 value greater the STOPR2 cutoff value of 0.005, the error mean square becomes larger
when this effect enters the model.
q
SS and R2 portion for Effects chosen for target: AMOUNT
Effect DF R2 SS EMS
-----------------------------------------------------------------------------
Var: LEISURE 1 0.4827 532690709 290726.1
Var: APPAREL 1 0.1344 148302146 215325.5
Class: LUXURY 1 0.0912 100683600 164118.4
Var: DOMESTIC 1 0.0489 53918925.2 136706.5
Var: DPM12 1 0.0101 11190481.5 131066.8
Class: KITCHEN 9 0.0086 9486959.66 126808.8
Class: STATECOD 54 0.0051 5659245.24 127435.3
Note: Note that the AOV16, GROUP, and two-way class interaction effects are not considered in the forward stepwise
regression. Including these effects may produce a better model, but it will also increase the execution time of the DMINE
procedure. To learn how to include these effects into the analysis, see Example 2.
Final ANOVA Table for the Target
The ANOVA table is divided into the following four columns:
Effect labels the source of variation as Model, Error, or Total.
q
DF lists the degrees of freedom for each source of variation.
q
R2 is the model R2, which is the ratio of the model sums of squares (SS) to the total sums of squares. In this
q
example, the inputs collectively explain 78.10% of the total variability in the target AMOUNT.
SS partitions the total target variation into portions that can be attributed to the model inputs and to error.
q
DMINE: Continous Target
The final ANOVA table for target: AMOUNT
Effect DF R2 SS
-------------------------------------------
Model 68 0.7810 861932067
Error 1897 241744735
Total 1965 1103676802
SS and R2 portion for Effects not chosen for Target
The final section lists the sums of squares and the R-square values for the effects that are not chosen in the final model.
SS and R2 portion for Effects not chosen for target: AMOUNT
Effect DF R2 SS
---------------------------------------------------------------
Var: HHAPPAR 1 0.0022 2405347
Var: TOWELS 1 0.0002 213272
Var: LINENS 1 0.0000 19937
Var: HOMEACC 1 0.0000 8551
Var: LAMPS 1 0.0003 351453
Var: PROMO7 1 0.0012 1342694
Var: MENSWARE 1 0.0027 3016894
Var: WAPPAR 1 0.0023 2489091
Var: BLANKETS 1 0.0005 517624
Var: PROMO13 1 0.0006 641861
Class: DISHES 9 0.0018 1991598
Var: COATS 1 0.0015 1706268
Copyright 2000 by SAS Institute Inc., Cary, NC, USA. All rights reserved.