Sie sind auf Seite 1von 12

GET

FILE='C:\Program
Files\IBM\SPSS\Statistics\21\Samples\English\car_sales.sav'.
DATASET NAME DataSet1 WINDOW=FRONT.
GET
FILE='C:\Program
Files\IBM\SPSS\Statistics\21\Samples\English\tree_credit.sav'.
DATASET NAME DataSet2 WINDOW=FRONT.
* Decision Tree.
TREE Credit_rating [n] BY Age [s] Income [o] Education [n] Car_loans [n]
/TREE DISPLAY=TOPDOWN NODES=STATISTICS BRANCHSTATISTICS=YES NODEDEFS=YES
SCALE=AUTO
/DEPCATEGORIES USEVALUES=[.00 1.00]
/PRINT MODELSUMMARY CLASSIFICATION RISK
/METHOD TYPE=CHAID
/GROWTHLIMIT MAXDEPTH=AUTO MINPARENTSIZE=100 MINCHILDSIZE=50
/VALIDATION TYPE=NONE OUTPUT=BOTHSAMPLES
/CHAID ALPHASPLIT=0.05 ALPHAMERGE=0.05 SPLITMERGED=NO CHISQUARE=PEARSON
CONVERGE=0.001 MAXITERATIONS=100 ADJUST=BONFERRONI INTERVALS=10
/COSTS EQUAL
/MISSING NOMINALMISSING=MISSING.

Classification Tree

Notes
Output Created

08-FEB-2013 16:17:08

Comments
C:\Program
Data

Files\IBM\SPSS\Statistics\21\Samples\E
nglish\tree_credit.sav

Input

Active Dataset

DataSet2

Filter

<none>

Weight

<none>

Split File

<none>

N of Rows in Working Data

2464

File
Missing Value Handling

Definition of Missing

Handling of user-defined missing values


of nominal independent variables
depends on the growing method.

Only cases with valid data for the


Cases Used

dependent variable and some or all


independent variables are used in
computing any statistics.
TREE Credit_rating [n] BY Age [s]
Income [o] Education [n] Car_loans [n]
/TREE DISPLAY=TOPDOWN
NODES=STATISTICS
BRANCHSTATISTICS=YES
NODEDEFS=YES SCALE=AUTO
/DEPCATEGORIES USEVALUES=[.00
1.00]
/PRINT MODELSUMMARY
CLASSIFICATION RISK
/METHOD TYPE=CHAID
/GROWTHLIMIT MAXDEPTH=AUTO
MINPARENTSIZE=100

Syntax

MINCHILDSIZE=50
/VALIDATION TYPE=NONE
OUTPUT=BOTHSAMPLES
/CHAID ALPHASPLIT=0.05
ALPHAMERGE=0.05
SPLITMERGED=NO
CHISQUARE=PEARSON
CONVERGE=0.001
MAXITERATIONS=100
ADJUST=BONFERRONI
INTERVALS=10
/COSTS EQUAL
/MISSING
NOMINALMISSING=MISSING.

Resources
Files Saved

Processor Time

00:00:01.47

Elapsed Time

00:00:01.77

Rules File

[DataSet2] C:\Program
Files\IBM\SPSS\Statistics\21\Samples\English\tree_credit.sav

Warnings
Gain summary Tables are not displayed because profits are undefined.
Target category gains tables are not displayed because target categories are undefined.

Model Summary

Specifications

Growing Method

CHAID

Dependent Variable

Credit rating

Independent Variables

Age, Income level, Education, Car loans

Validation

None

Maximum Tree Depth

Minimum Cases in Parent

100

Node
Minimum Cases in Child

50

Node
Independent Variables

Income level, Car loans, Age

Included
Results

Number of Nodes

28

Number of Terminal Nodes

17

Depth

Risk
Estimate
.204

Std. Error
.008

Growing Method: CHAID


Dependent Variable: Credit
rating

Classification
Observed

Predicted
Bad

Good

Percent Correct

Bad

723

297

70.9%

Good

206

1238

85.7%

37.7%

62.3%

79.6%

Overall Percentage
Growing Method: CHAID

Dependent Variable: Credit rating

GET
FILE='C:\Program Files\IBM\SPSS\Statistics\21\Samples\English\bankloan.sav'.
DATASET NAME DataSet3 WINDOW=FRONT.
* Decision Tree.
TREE default [n] BY age [s] ed [o] employ [s] address [s] income [s] debtinc
[s] creddebt [s] othdebt [s]
/TREE DISPLAY=TOPDOWN NODES=STATISTICS BRANCHSTATISTICS=YES NODEDEFS=YES
SCALE=AUTO
/DEPCATEGORIES USEVALUES=[0 1]
/PRINT MODELSUMMARY CLASSIFICATION RISK
/METHOD TYPE=CHAID
/GROWTHLIMIT MAXDEPTH=AUTO MINPARENTSIZE=100 MINCHILDSIZE=50
/VALIDATION TYPE=NONE OUTPUT=BOTHSAMPLES
/CHAID ALPHASPLIT=0.05 ALPHAMERGE=0.05 SPLITMERGED=NO CHISQUARE=PEARSON
CONVERGE=0.001 MAXITERATIONS=100 ADJUST=BONFERRONI INTERVALS=10
/COSTS EQUAL.

Classification Tree

Notes
Output Created
Comments

08-FEB-2013 16:18:39

C:\Program
Data

Files\IBM\SPSS\Statistics\21\Samples\E
nglish\bankloan.sav

Input

Active Dataset

DataSet3

File Label

Bank Loan Default

Filter

<none>

Weight

<none>

Split File

<none>

N of Rows in Working Data

850

File
Handling of user-defined missing values
Definition of Missing

of nominal independent variables


depends on the growing method.

Missing Value Handling

Only cases with valid data for the


Cases Used

dependent variable and some or all


independent variables are used in
computing any statistics.

TREE default [n] BY age [s] ed [o]


employ [s] address [s] income [s]
debtinc [s] creddebt [s] othdebt [s]
/TREE DISPLAY=TOPDOWN
NODES=STATISTICS
BRANCHSTATISTICS=YES
NODEDEFS=YES SCALE=AUTO
/DEPCATEGORIES USEVALUES=[0
1]
/PRINT MODELSUMMARY
CLASSIFICATION RISK
/METHOD TYPE=CHAID
/GROWTHLIMIT MAXDEPTH=AUTO

Syntax

MINPARENTSIZE=100
MINCHILDSIZE=50
/VALIDATION TYPE=NONE
OUTPUT=BOTHSAMPLES
/CHAID ALPHASPLIT=0.05
ALPHAMERGE=0.05
SPLITMERGED=NO
CHISQUARE=PEARSON
CONVERGE=0.001
MAXITERATIONS=100
ADJUST=BONFERRONI
INTERVALS=10
/COSTS EQUAL.

Resources
Files Saved

Processor Time

00:00:00.31

Elapsed Time

00:00:00.27

Rules File

[DataSet3] C:\Program
Files\IBM\SPSS\Statistics\21\Samples\English\bankloan.sav

Warnings
Gain summary Tables are not displayed because profits are undefined.
Target category gains tables are not displayed because target categories are undefined.

Model Summary
Growing Method

CHAID

Dependent Variable

Previously defaulted
Age in years, Level of education, Years with
current employer, Years at current address,

Independent Variables

Household income in thousands, Debt to income


ratio (x100), Credit card debt in thousands, Other
debt in thousands

Specifications
Validation

None

Maximum Tree Depth

Minimum Cases in Parent

100

Node
Minimum Cases in Child

50

Node

Results

Independent Variables

Debt to income ratio (x100), Years with current

Included

employer, Years at current address

Number of Nodes

13

Number of Terminal Nodes

Depth

Risk
Estimate

Std. Error

.220

.016

Growing Method: CHAID


Dependent Variable:
Previously defaulted

Classification
Observed

Predicted
No

Yes

Percent Correct

No

488

29

94.4%

Yes

125

58

31.7%

87.6%

12.4%

78.0%

Overall Percentage
Growing Method: CHAID

Dependent Variable: Previously defaulted

* Decision Tree.
TREE default [n] BY age [s] ed [o] employ [s] address [s] income [s] debtinc
[s] creddebt [s] othdebt [s]
/TREE DISPLAY=TOPDOWN NODES=STATISTICS BRANCHSTATISTICS=YES NODEDEFS=YES
SCALE=AUTO
/DEPCATEGORIES USEVALUES=[0 1]
/PRINT MODELSUMMARY CLASSIFICATION RISK
/METHOD TYPE=CHAID
/GROWTHLIMIT MAXDEPTH=AUTO MINPARENTSIZE=100 MINCHILDSIZE=50
/VALIDATION TYPE=NONE OUTPUT=BOTHSAMPLES
/CHAID ALPHASPLIT=0.05 ALPHAMERGE=0.05 SPLITMERGED=NO CHISQUARE=PEARSON
CONVERGE=0.001 MAXITERATIONS=100 ADJUST=BONFERRONI INTERVALS=10
/COSTS EQUAL.

Classification Tree

Notes
Output Created
Comments

08-FEB-2013 16:21:17

C:\Program
Data

Files\IBM\SPSS\Statistics\21\Samples\E
nglish\bankloan.sav

Input

Active Dataset

DataSet3

File Label

Bank Loan Default

Filter

<none>

Weight

<none>

Split File

<none>

N of Rows in Working Data

850

File
Handling of user-defined missing values
Definition of Missing

of nominal independent variables


depends on the growing method.

Missing Value Handling

Only cases with valid data for the


Cases Used

dependent variable and some or all


independent variables are used in
computing any statistics.

TREE default [n] BY age [s] ed [o]


employ [s] address [s] income [s]
debtinc [s] creddebt [s] othdebt [s]
/TREE DISPLAY=TOPDOWN
NODES=STATISTICS
BRANCHSTATISTICS=YES
NODEDEFS=YES SCALE=AUTO
/DEPCATEGORIES USEVALUES=[0
1]
/PRINT MODELSUMMARY
CLASSIFICATION RISK
/METHOD TYPE=CHAID
/GROWTHLIMIT MAXDEPTH=AUTO

Syntax

MINPARENTSIZE=100
MINCHILDSIZE=50
/VALIDATION TYPE=NONE
OUTPUT=BOTHSAMPLES
/CHAID ALPHASPLIT=0.05
ALPHAMERGE=0.05
SPLITMERGED=NO
CHISQUARE=PEARSON
CONVERGE=0.001
MAXITERATIONS=100
ADJUST=BONFERRONI
INTERVALS=10
/COSTS EQUAL.

Resources
Files Saved

Processor Time

00:00:00.28

Elapsed Time

00:00:00.25

Rules File

[DataSet3] C:\Program
Files\IBM\SPSS\Statistics\21\Samples\English\bankloan.sav

Warnings
Gain summary Tables are not displayed because profits are undefined.
Target category gains tables are not displayed because target categories are undefined.

Model Summary
Growing Method

CHAID

Dependent Variable

Previously defaulted
Age in years, Level of education, Years with
current employer, Years at current address,

Independent Variables

Household income in thousands, Debt to income


ratio (x100), Credit card debt in thousands, Other
debt in thousands

Specifications
Validation

None

Maximum Tree Depth

Minimum Cases in Parent

100

Node
Minimum Cases in Child

50

Node

Results

Independent Variables

Debt to income ratio (x100), Years with current

Included

employer, Years at current address

Number of Nodes

13

Number of Terminal Nodes

Depth

Risk
Estimate

Std. Error

.220

.016

Growing Method: CHAID


Dependent Variable:
Previously defaulted

Classification
Observed

Predicted
No

Yes

Percent Correct

No

488

29

94.4%

Yes

125

58

31.7%

87.6%

12.4%

78.0%

Overall Percentage
Growing Method: CHAID

Dependent Variable: Previously defaulted

Das könnte Ihnen auch gefallen