Hello I am building a credit score model What is acceptable difference in Gini between in-sample( train data) to out of time ? For example; Gini in in-sample is 80% and Gini in out of time data is 82%. Is it good or bad? I afraid that 2% difference means model is not good? ... View more

Hello I am building a credit score model. I almost finished build the model and I calculated Gini . The model is developed based on data from Jan 2023 and then check default in Feb 2023 till Jan 2024( 12 months). Note that the data divided to train (70%) and test (30%). I was asked to do the following- Find customers with very bad elements that are "Indeterminate" (in Jan 2023) but still not in default (in Jan 2023). It means that these customers are very  close to be defaulted and already have days past due (but still not enough to be in default) . I read that these type of customers are called "Indeterminate". My question- The task is to find if the model has good separation power for the "good " customers (customers who are not Indeterminate). The idea is that for customers who are not indeterminate it is more difficult to predict default. What is the way to check it( if the modelhas good separation power for the "good " customers)? ... View more

Hello I want to create a new variable that classify each customerID into a group between 1 and 100 by percentiles of salary amount. Please note that I want that the percentiles will be calculated only on cusomers with positive salary (customers with salary zero will not influence on the percentiles calculation). So for example: customers with salary 0 be in group 0 customers with salary greater than 0 and lower equal to percentile 1 will be in group 1 customers with salary greater than percentile 1 and lower equal to percentile 2 will be in group 2 and so on customers with salary greater than percentile 99 and lower equal to percentile 100( max) will be in group 100   What is the way to do it please       ... View more

Hello I have a macro called RRR %Macro RRR(V_months); code here %mend RRR; Usually I run the macro by typing the months as you see here %RRR_STEP1(V_months=2211+2210+2209+2208+2207+2206) %RRR_STEP1(V_months=2210+2209+2208+2207+2206+2205) %RRR_STEP1(V_months=2209+2208+2207+2206+2205+2204) %RRR_STEP1(V_months=2208+2207+2206+2205+2204+2203) %RRR_STEP1(V_months=2207+2206+2205+2204+2203+2202) %RRR_STEP1(V_months=2206+2205+2204+2203+2202+2201) %RRR_STEP1(V_months=2205+2204+2203+2202+2201+2112) %RRR_STEP1(V_months=2204+2203+2202+2201+2112+2111) %RRR_STEP1(V_months=2203+2202+2201+2112+2111+2110) %RRR_STEP1(V_months=2202+2201+2112+2111+2110+2109) %RRR_STEP1(V_months=2201+2112+2111+2110+2109+2108) %RRR_STEP1(V_months=2112+2111+2110+2109+2108+2107) But it is not so comfortable to type so many dates manually. Let's say that there is a macro var called Vector_CAT that automatically calculate the macro dates that need to run. In this case this macro var get value : 2306+2305+2304+2303+2302+2301,2307+2306+2305+2304+2303+2302,2308+2307+2306+2305+2304+2303,2309+2308+2307+2306+2305+2304,2310+2309+23 08+2307+2306+2305   I want to run the macro multiple times  I try this code but I am not sure it is correct. What do you think? is it correct? Why  should I write the delimiter (comma)  like this :   %str(,)    ? Is there another good way to run the macro automatically from the macro varaible components? %macro run_step1; %do i=1 %to &n_run_step1; %let mm = %scan(%quote(&Vector_CAT), &i, %str(,)); %RRR_STEP1(&mm); %end; %mend run_step1; %run_step1;   ... View more