Does anyone know how SPSS resets the actual the mean and standard deviation when you replace outliers with for example 3 St Deviations in the Auto Data Prep module. The syntax that shows the procedure actual lowers the actual mean to another value before it applies the 3 STD adjustment. Looks like some type of trimming or other adjustment to get to a different mean & SDev. The mean and standard deviation in the outlier adjustment code does not match the original means and standard deviations for the variables. I am looking for what formula is used to get to the different mean & StDev they use in the syntax.
I have to explain to a client how it is calculated. Please help if anyone can. Or direct me to someone who might know.
Below is the syntax, but the total mean and SD were higher.
*Interactive Data Preparation.
COMPUTE #MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT_Outlier = $SYSMIS.
DO IF (MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT < 1617.62946991648-3508.74610065027*3).
COMPUTE #MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT_Outlier = 1617.62946991648-3508.74610065027*3.
ELSE IF (MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT > 1617.62946991648+3508.74610065027*3).
COMPUTE #MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT_Outlier = 1617.62946991648+3508.74610065027*3.
ELSE.
COMPUTE #MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT_Outlier = MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT.
END IF.
COMPUTE MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT_transformedTest = (((1/3176.06783869562)*(#MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT_Outlier-1669.24263701691))+0).
VARIABLE ROLE
/NONE MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT
/INPUT MBR_SAV_SUMMARY_TOTAL_TRXN_CREDIT_60DAYS_AMT_transformedTest.
EXECUTE.
------------------------------
MARTHA REA
------------------------------