Hello,

Can you please help me to understand this Dmine regression?

This is a fraudulent claim dataset, and I need to examine the results.

Thank you very much

9 10 11 12 Variable Summary 13 14 Measurement Frequency 15 Role Level Count 16 17 ID INTERVAL 18 INPUT INTERVAL 19 INPUT NOMINAL 12 20 REJECTED NOMINAL 21 TARGET BINARY 22 23 24 25 26 Output Table Attributes 27 Variables Measurement Level Label Drop Variable Name Role Order Annual Premium No Annual Premium Input Interval No Claim Amount Input Interval Claim Amount Claim Cause No Claim Cause Input Nominal Claim Date No Claim Date Input Nominal Claim Report Type No Claim Report Type Input Nominal No Claimant Number ID Interval Education No Education Input Nominal Input Nominal Employment Status No Employment Status No Fraudulent Claim Target Binary Fraudulent Claim Gender No Gender Input Nominal No Income Input Interval Income Location No Location Input Nominal Marital Status No Marital Status Input Nominal Monthly Premium No Monthly Premium Input Interval Input Interval Months Since Last Claim No Months Since Last Claim No Months Since Policy Inception Input Interval Months Since Policy Inception Outstanding Balance No Outstanding Balance Input Interval Rejected Nominal State No State State Code No State Code Input Nominal Vehicle Class No Vehicle Class Input Nominal Vehicle Model Vehicle Model Input Nominal No No Vehicle Size Input Nominal Vehicle SizeResults - Node: Dmine Regression Diagram: Data Preparation and Analysis File Edit View Window 90604 Score Rankings Overlay: Fraudulent_Claim O X Fit Statistics Cumulative Lift Target Target Label Fit Statistics Statistics Label Train Validation ASE 0.05246 5- Fraudulent Claim Fraudulent Claim Average Squared Error 0.05711 Fraudulent Claim Fraudulent Claim DIV Divisor for ASE 5996 400 Fraudulent Claim Fraudulent Claim MAX Maximum Absolute Error 0.988636 0.99662 Fraudulent Claim Fraudulent Claim NOBS Sum of Frequencies 2998 200 Fraudulent Claim Fraudulent Claim RASE Root Average Squared 0.229042 0.23899 Fraudulent Claim Fraudulent Claim Sum of Squared Errors 314.5512 228.705 Fraudulent Claim Fraudulent Claim DISP Frequency of Classified. 2998 200 4- Fraudulent Claim Fraudulent Claim MISC Misclassification Rate 0.061374 0.06443 Fraudulent Claim Fraudulent Claim WRONG Number of Wrong Clas.. 184 Cumulative Lift 2 - 20 40 60 80 100 Depth TRAIN -VALIDATE Il Effects in the Mode -OX Output 13 Measurement 0.020 14 Frequency 15 Role Level Count 16 17 INPUT INTERVAL 0.015 18 INPUT NOMINAL 19 REJECTED NOMINAL 20 TARGET BINARY Sequential R-Square 21 0.010- 22 23 24 0.005 Model Events 26 Number Measurement of 0.000 - Target Event Level Levels Order Label Gender Fraudulent_Claim y BINARY 2 Descending Fraudulent_Claim G Vehicle_Class G Claim_Cause AOV16_Claim_Amount AOV16 Months_Since_Last_Claim Predicted and decision variables Type Variable Label TARGET Fraudulent_Claim Fraudulent_Claim PREDICTED _Fraudulent_ClaimY Predicted: Fraudulent_Claim=Y RESIDUAL _Fraudulent_ClaimY Residual: Fraudulent_Claim=Y PREDICTED P_Fraudulent_ClaimN Predicted: Fraudulent_Clain-N RESIDUAL R_Fraudulent_ClaimN Residual: Fraudulent_Claim=N Effect FROM _Fraudulent_Claim From: Fraudulent_ClaimOutput 10 11 12 Variable Summary 13 14 Measurement Frequency 15 Role Level Count 16 17 INPUT INTERVAL 18 INPUT NOMINAL 12 19 REJECTED NOMINAL 20 TARGET BINARY 21 22 23 24 25 Model Events 26 27 Number 26 Heasurement 29 Target Event Level Levels Order Label 30 31 Fraudulent_Claim Y BINARY 2 Descending Fraudulent_Claim 32 33 34 35 36 Predicted and decision variables 37 38 Type Variable Label 39 40 TARGET Fraudulent_Claim Fraudulent_Claim 41 PREDICTED P_Fraudulent_ClaimY Predicted: Fraudulent_Claim=Y 42 RESIDUAL R_Fraudulent_ClaimY Residual: Fraudulent_Claim-Y 43 PREDICTED P_Fraudulent_ClaimN Predicted: Fraudulent_Clain=N 44 RESIDUAL R_Fraudulent_ClaimN Residual: Fraudulent_Claim=N 45 FROM F_Fraudulent_Claim From: Fraudulent_Claim 46 INTO I_Fraudulent_Claim Into: Fraudulent_Claim 47 48 40Results - Node: Dmine Regression Diagram: Data Preparation and Analysis File Edit View Window Output 226 |227 228 229 The DMINE Procedure 230 231 Effects Chosen for Target: Fraudulent_Claim 232 233 Sun of Error Mean 234 Effect R-Square F Value P-Value Squares Square 235 236 Group: Vehicle_Class 0. 022049 33. 762833 <. class: gender . group: claim_cause aqv16: months_since_last_claim adv16: claim_amount the dmine procedure final anova table for target: fraudulent_claim sum of effect df r-square squares model error total estimating logistic iter alpha beta classification cutoff="0." accuracy="93.86" predicted observed y n missing node: regression diagram: data preparation and analysis file edit view window output fit statistics target="Fraudulent_Claim" label="Fraudulent_Claim" train validation ase average squared div_ divisor max maximum absolute nobs frequencies rase_ root _sse errors disf_ frequency classified cases hisc_ hisclassification rate wrong number classifications role="TRAIN" variable="Fraudulent_Claim" outcome percentage count here variable-fraudulent_claim label-fraudulent_claim event false true negative positive assessment score rankings mean cumulative posterior depth gain lift response observations probability distribution ol range events nonevents d. jubanpeenno>