The arboretum procedure

       Group: KITCHEN*RACE                          6      0.0210

       AOV16: TOWELS                               10      0.0208

       AOV16: OUTDOOR                              11      0.0201

       Class: GENDER*KITCHEN                       16      0.0197    R2 < MINR2

       Group: GENDER*KITCHEN                        5      0.0195    R2 < MINR2

       Class: LUXURY*DISHES                        15      0.0193    R2 < MINR2

       Group: LUXURY*DISHES                         5      0.0190    R2 < MINR2

       Class: APRTMNT*KITCHEN                      15      0.0189    R2 < MINR2

       Class: SNGLMOM*KITCHEN                      15      0.0187    R2 < MINR2

       Group: APRTMNT*KITCHEN                       4      0.0186    R2 < MINR2

       AOV16: HHAPPAR                              13      0.0185    R2 < MINR2

       AOV16: JEWELRY                              10      0.0185    R2 < MINR2

       Group: SNGLMOM*KITCHEN                       4      0.0184    R2 < MINR2

       Class: MOBILE*KITCHEN                       15      0.0184    R2 < MINR2

       Group: MOBILE*KITCHEN                        5      0.0182    R2 < MINR2

       AOV16: LAMPS                                12      0.0178    R2 < MINR2

       Class: KITCHEN                               9      0.0174    R2 < MINR2

       AOV16: LINENS                               13      0.0173    R2 < MINR2

       AOV16: PROMO13                              15      0.0172    R2 < MINR2

       Group: KITCHEN                               3      0.0171    R2 < MINR2

       AOV16: BLANKETS                             11      0.0169    R2 < MINR2

       Class: LUXURY*ORIGIN                        11      0.0168    R2 < MINR2

       Group: LUXURY*ORIGIN                         5      0.0167    R2 < MINR2

       Var:   OUTDOOR                               1      0.0166    R2 < MINR2

       Class: DISHES*HEAT                          24      0.0165    R2 < MINR2

       Additional Effects Are Not Listed

SS and R2 Portion for Effects Chosen for 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 R-square 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 KITCHEN*STATECOD interaction as "10.45%

of the variation in the target PURCHASE is explained by its linear relationship with this effect". The R2 statistic for

NTITLE*STATECOD indicates that this two-factor interaction accounts for an additional 6.38% 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.

q   

  

                                DMINE: Binary Target

  Effect                                  DF        R2            SS           EMS

  ----------------------------------------------------------------------------------

  Class: KITCHEN*STATECOD                197    0.1045      51.35546     0.2488769

  Class: NTITLE*STATECOD                 132    0.0683      33.55930     0.2484444

  Class: STATECOD*ORIGIN                 106    0.0660      32.43976     0.2444544

  Class: DISHES*STATECOD                 100    0.0473      23.21811     0.2453127

---

  Class: STATECOD*EDLEVEL                 83    0.0437      21.49171     0.2444732

  AOV16: FREQUENT                          9    0.0332      16.30766     0.2339296

  Class: STATECOD*HEAT                    73    0.0369      18.15072     0.2330807

  Class: STATECOD*NUMCARS                 55    0.0237      11.66560     0.2340343

  Class: STATECOD*RACE                    40    0.0228      11.18392     0.2324765

  Class: LUXURY*STATECOD                  37    0.0188       9.22242     0.2319286

  Class: MARITAL*STATECOD                 39    0.0172       8.45564     0.2324675

  Group: TMKTORD*STATECOD                  8    0.0093       4.55481     0.2299859

  Var:   RECENCY                           1    0.0066       3.25424     0.2271985

The 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 example,

the selected inputs collectively explain 49.83% of the total variability in the target PURCHASE.

q   

SS partitions the total target variation into portions that can be attributed to the model inputs and to error.

q   

 

                  The final ANOVA table for target: PURCHASE

                  Effect           DF        R2            SS

                  -------------------------------------------

                  Model           880    0.4983     244.85936

                  Error          1085               246.51043

                  Total          1965               491.36979

SS and R2 portion for Effects Not Chosen for the Target: PURCHASE

 

         SS and R2 portion for Effects not chosen for target: PURCHASE

         Effect                             DF        R2            SS

        ---------------------------------------------------------------

         Class: GENDER*STATECOD              0    0.0000       0.00000

         Var:   FREQUENT                     1    0.0010       0.50991

         Var:   DOMESTIC                     1    0.0012       0.57574

Estimating Logistic

When the target is binary, predicted values or SUPERX's are computed from the forward stepwise regression. The SUPERX's

are then grouped into 256 equally spaced intervals, which are used as the independent variable in a final logistic regression

analysis. The logistic regression helps you decide the cutoff of the binary response. Since there is one input, only two

parameters are estimated (the intercept and the slope).

The first table shows the iteration history for estimating the intercept (alpha) and the slope (beta) for the approximate logistic

regression.

The second table contains the predicted values, which are bucketed into the 256 equally sized sub-intervals. The table

---