cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
blue34
‎04-02-2020
  • 17 Posts
  • 0 Likes Given
  • 0 Solutions
  • 0 Likes Received

Re: Statistical test

by Calcite | Level 5 in SAS Academy for Data Science
‎04-02-2020 11:00 AM
‎04-02-2020 11:00 AM
Also, I don't have separate models. I have one model and tested on two different samples. I stated as "I have one classification model and tested on two samples" ... View more

Re: Statistical test

by Calcite | Level 5 in SAS Academy for Data Science
‎04-02-2020 10:28 AM
‎04-02-2020 10:28 AM
I thought this is very simple Z test I should apply, Z test for precision of female and male and check the p value. Also Z test for recall of female and male and check the p value whether it is statistically same or not. I am not sure if we need this complication. What do you think? ... View more

Statistical test

by Calcite | Level 5 in SAS Academy for Data Science
‎03-31-2020 01:43 PM
‎03-31-2020 01:43 PM
I have one classification model and tested on two samples. I have two confusion matrix for both samples, one of them is for female, other is male only, both samples are not paired. I calculated precision, recall, F1 score for both and how can I statistically compare precision, recall, F1 score between female/male? Which test i can use ? ... View more

Decision Tree Model Development / Validation

by Calcite | Level 5 in SAS Data Science
‎09-24-2018 09:14 PM
‎09-24-2018 09:14 PM
Hi All,   I have been working on one Decision Tree  on SAS EM. I created Validation / Train dataset for hold-out validation and run decision tree model. I will have two questions, appreciate if you could help.   1. How can I make sure train and validation data look reasonable besides checking GINI, ROC? Any other thoughts I have to look at?   2. Do you think, does it make sense to do out-of-time validation in Decision Tree Model ? i know we should do this type of validation in credit scoring, but not sure if it is worth or appropriate for Decision Tree Models.   All feedback welcome, appreciate it! Please feel free to add if you have any other thoughts regarding DT development and validation.   ... View more

Re: Balance flag looking two months back

by Calcite | Level 5 in SAS Programming
‎07-08-2018 05:58 PM
‎07-08-2018 05:58 PM
I just wrote this for 18 months back flag creation, will try this code. %macro test(); data want; set account_fdr_hist2; by account_fdr time_id; if first.account_fdr then recnum=1; else recnum + 1; %do i=1 %to 18; flag_back&i = (lag&i(balance) <> 0 ) and (recnum > &i); flag = max(flag_back&i); /*drop recnum flag_back1 flag_back2;*/ run; %mend; %test; ... View more

Re: Balance flag looking two months back

by Calcite | Level 5 in SAS Programming
‎07-08-2018 05:42 PM
‎07-08-2018 05:42 PM
Thank you very much!!!!!   This worked on SAS. I would like to ask one question, if I want to create flag back to 18 months, how can I do this? ... View more

Re: Balance flag looking two months back

by Calcite | Level 5 in SAS Programming
‎07-08-2018 05:22 PM
‎07-08-2018 05:22 PM
When I run this code, I am getting for 5 account number in Feb 2018 as 0 flag. However, I want to see for this account as 1 since it has Jan-2018 Balance already. ... View more

Re: Balance flag looking two months back

by Calcite | Level 5 in SAS Programming
‎07-08-2018 05:17 PM
‎07-08-2018 05:17 PM
Thank you, if I wanna look at back to 18 months instead of 2 months, do you have another solution using loop? ... View more

Balance flag looking two months back

by Calcite | Level 5 in SAS Programming
‎07-08-2018 04:35 PM
‎07-08-2018 04:35 PM
Hi,     I am trying to write a code for below example of data: For each time ID of accounts, that have a balance greater than "0" in any one of the two previous months (not current)  how can we create a flag ?     time id account number balance Jan-18 5 50 Feb-18 5 50 Mar-18 5 50 Apr-18 5 50 May-18 5 30 Jun-18 5 . Jul-18 5 . Aug-18 5 40 Sep-18 5 34 Jan-18 8 96 Feb-18 8 67 Mar-18 8 78 Apr-18 8 12 ... View more

Re: Data Cleaning Logic

by Calcite | Level 5 in SAS Programming
‎01-31-2018 11:37 AM
‎01-31-2018 11:37 AM
Thank you very much, this is very helpful. I notice that I have this below conversion, so I have duplicate account, quarters now. How I can deal with you code for making data quarterly?   transform the data into quarterly end in month 2,5,8,11 12,1,2-Q1 3,4,5 -Q2 6,7,8 -Q3 9,10,11 -Q4     year    quarter    account    status    default   1       Q1         230        AC         0     1       Q2         230        CL         0     1       Q3         230        CL         0     1       Q1         231        AC         0     1       Q2         231        CL         0     1       Q3         231        CL         0     1       Q4         231        CL         0     2       Q1         231        CL         0     2       Q2         231        CL         1   ... View more

Data Cleaning Logic

by Calcite | Level 5 in SAS Programming
‎01-31-2018 02:04 AM
‎01-31-2018 02:04 AM
Hi,   I am trying to clean my data based on below items. Could you help me to write code for this?    From "AC" to "CL" status, but never be identified as default=1 where accounts stay two more months and after that I should remove all accounts. Below example my code should delete Q4 row. quarter account status DEFault   Q1 230 AC 0   Q2 230 CL 0 Keep Q3 230 CL 0 Keep Q4 230 CL 0 Remove Q1 230 CL 0 Remove Q2 230 CL 0 Remove Q3 230 CL 0 Remove Q4 230 CL 0 Remove   Be identified as default in later days. If account closed but have default later days, we should keep those. But after default flag where we see (below example Q2),  Q3 and Q4 quarters should remove. quarter account status DEFault     Q1 230 AC 0     Q2 230 CL 0 Keep   Q2 230 CL 0 Keep   Q3 230 CL 0 Keep   Q1 230 CL 0 Keep   Q2 230 CL 1 Keep   Q3 230 CL 0 Remve Already default Q4 230 CL 0 Remove Already default   ... View more

Re: Arrays and Do loops

by Calcite | Level 5 in SAS Programming
‎09-30-2017 10:14 PM
‎09-30-2017 10:14 PM
Thank you very much again!! ... View more

Re: Arrays and Do loops

by Calcite | Level 5 in SAS Programming
‎09-30-2017 10:00 PM
‎09-30-2017 10:00 PM
I just did, Sounds great! Thank you! But how about instead of putting prefix Q, I want to create fields as Q1, Q2, Q3 to Q20 ? ... View more

Re: Arrays and Do loops

by Calcite | Level 5 in SAS Programming
‎09-30-2017 09:45 PM
‎09-30-2017 09:45 PM
I don`t think this will work, because I want amount into 20 fields after I transpose. Am i missing something   Amount will be A1-A5 for each performance_date each account   acccount    A1   A2  A3  A4 A5 1 2 ... View more

Re: Arrays and Do loops

by Calcite | Level 5 in SAS Programming
‎09-30-2017 09:15 PM
‎09-30-2017 09:15 PM
After create this ID I will transpose this data, that is why I am trying to create that ID field.   My final dataset after transpose will be based on account number:   acccount    A1   A2  A3  A4 A5 1 2   This is my full code and want to automatize this:   data data set data; length quarter $ 11; if performance_dt = '31dec2016'd then quarter="A0"; if performance_dt = '31mar2017'd then quarter="A1"; if performance_dt = '30jun2017'd then quarter="A2"; if performance_dt = '30sep2017'd then quarter="A3"; if performance_dt = '31dec2017'd then quarter="A4"; if performance_dt = '31mar2018'd then quarter="A5"; if performance_dt = '30jun2018'd then quarter="A6"; if performance_dt = '30sep2018'd then quarter="A7"; keep account quarter amount; run; proc sort data=data; by account ;run; proc transpose data=data out=data; by account; id quarter; run;       ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎04-02-2020 02:21 PM
Quick links