REPAIR_NOTICE_DATE_2
REPAIR_TYPE_2
REPAIR_DEBT_2_WITHHELD
DEBT_PAID_AFTER_COMPLETION_2
REPAIR_TYPE_3
REPAIR_DEBT_WITHHELD_3
DEBT_PAID_AFTER_COMPLETION_3
REPAIR_NOTICE_DATE_3
TOTAL_REPAIR_DEBT_OUTSTANDING
CONTACTED
CONTACTED_BY_EMAIL
CONTACTED_BY_HOME_VISIT
CONTACTED_BY_LETTER
CONTACTED_BY_LARGE_PRINT
EMAIL_SUCCESS
HOME_VISIT_SUCCESS
LETTER_SUCCESS
LARGE_PRINT_SUCCESS
1.
On the next screen,
Summary of Modeling Parameters,
select Generate
2.
The
Model Overview
screen then appears
3.
View the KI and KR values
4.
Go to the Next and view various Display Views
It is important to display
Clusters Summary
as it is necessary to know which Clusters are "good"
ones in the sense that people grouped within them are likely to respond favorably to a Home
Visit and will then make their payment. Let the mouse hover over the cluster circles on the right
hand side until the Cluster Number appears (like to be 2,3, and 5). Another way to determine the
"good" clusters is to display the
Cluster Profiles
and note which ones have a high percentage of
1. The same cluster numbers should be determined: 2, 3, and 5.
5.
On the Using the Model screen, select RUN then Apply Model
6.
On the
Applying the Model
screen, enter VW_PREDICT_0 - these are all of the non-payers to be evaluated
7.
Leave the
Generation Options
as they are
8.
In the
Results Generated by the Model
section, set the
Data Type
to
Text Files
, then specify a directory and
a file name on your computer where you want the output file to go.
9.
When the model was applied successfully, select View Output
Note that this same output is in the output file specified previously.
10.
To save this Model, select
Save /Export
– Save Model
on the
Using the Model
screen
11.
Fill in the Model Name and Description, set the Data Type to Text File and specify the folder and filename
where it is to be saved.
If other Contact Methods are to be predicted, substitute VW_PREDICT_1_EMAIL,
VW_PREDICT_1_LETTER, or VW_PREDICT_1_LARGE_PRINT for
VW_PREDICT_1_HOME_VISIT in the
Select a Data Source
when building the Cluster Model.
