About danirecchia

danirecchia · ‎02-08-2013

Many Thanks!!

danirecchia · ‎02-06-2013

Hello, I am pretty new by web report and I want to know if the refresh bottom could actually make the sas code run again. I made a project in EG and exported the report to web studio. I can see it there but I would like to web report run the project again and display the new values when I refresh the report (refresh bottom). When I refreshed the report now this is not happening, is it possible to have something like that? Thanks! Daniela

danirecchia · ‎01-25-2013

Hello Anca, it worked! Tahnk you very much. Kind regards, Daniela

danirecchia · ‎01-24-2013

Hello, on my dataset I have a column in date format such as "12Sep07:0:0:0" and I want to convert this in to "Sep07" I tried put(mmyyyyp8.) and another variations but without success. Do you have any idea how to convert this variable? Many thanks! Daniela

danirecchia · ‎09-13-2012

Hey tarek03, yep, the c-statistic is a very used measure and I tought on this comparison but you see, if I just look for the values of the c-statistic I am not able to compare them. But now, whith the ROCCONTRAST I know what to do. But I also think that a very interesting test would be the one, which I could compare the misclassification rate from the models, maybe this could be more specific on the discrimination power of the model. But for this case I don't know yet what to do.

danirecchia · ‎09-13-2012

Hello Doc@Duke, thank you very much for the explanation! I see now, what should I do.

danirecchia · ‎09-13-2012

Hello Rick, as I saw in the example that you sent, the ROCCONTRAST compares the ROC Curves from different the 3 different models with the 3 variables. But you see, for me originally the data is the same, but to adjust the model with categorized data I had to do some modifications and I can't see how should I use this option to compare the curves, since not only the data (structural - variables in classes) but also the chosen variables, I would like to compare both final models from row - and categorized data.

danirecchia · ‎09-13-2012

Hello everyone, I have one dataset and I want to model the probability of default from a loan. I used the logistic regression for it but with the data in different structures. I do use the same (original) data, but one model I used the row data and for another model I used the categorized data (I built all the categories) and now I want to be able to compare the two models in order to say if the structure of the data has some influence on the results. I tought I could maybe compare their misclassification rate, but how could I build such a test to make this comparison, since here I have only one value (the misclassification rate) for each model? Do you have some idea? Thanks

danirecchia · ‎09-06-2012

Hey Steve! I found it!!! And now, everything is working very well. Many thanks.

danirecchia · ‎09-06-2012

Hey Steve, well I didn't look at all combinations, because when I try to create a table with all the variables SAS tells me "ERROR: The requested table is too large to process." Maybe I don't know how to ask this for SAS. How should I do it?

danirecchia · ‎09-06-2012

Hey StatDave@sas thank you for your suggestions. Unfortunately none of them could help me. I tried everything, firth, forward, stepwise, exact logistic, observation of the estimates and standadr deviation and so on. There is no zero count in any of my categories. I then built the model inserting one by one and now I know which variables are causing the problem (15 variables) but I dond't understand why. I just can't let all this variables out of my model without a justification. Can you imagine another solution? Thanks

danirecchia · ‎08-21-2012

Hello dear users, I am wondering if I cant just exclude one group from my class variable by the logistic regression I mean, if I have the variable age: proc format; value agegroup 18-25 = 'age1' 25-35 = 'age2' 35-high= 'age3'; run; data test; set scoring; retain Age_cat; Age_cat = put(age,agegroup.); run; proc logistic data = test; class Age_cat; model y=Age_cat + x1; run; If I see that for example the age2 is no significant, can I just delete this group from the logistic procedure? Thank you proc logistic with grouped variables

danirecchia · ‎08-10-2012

Hello Steve, thank you for your answer. I create the variables so : proc format; value AVG_DELAY_L24Mgroup low-0.5 = 'delay1' 0.5-high = 'delay2';run; PROC FREQ data=score2; TABLES AVG_DELAY_L24M*bad_client; format AVG_DELAY_L24M AVG_DELAY_L24Mgroup.;RUN; then I receive the table: Category Good Bad delay1 437 376 delay2 40 239 another variable: proc format; value DUR_CUST_RELgroup low-500 = 'dur_cust_rel1' 500 - 1000 = 'dur_cust_rel2' 1000 - 3000 = 'dur_cust_rel3' 3000 - 5000 = 'dur_cust_rel4' 5000 - high = 'dur_cust_rel5'; run; table: Category Good Bad dur_cust_rel1 61 66 dur_cust_rel2 51 108 dur_cust_rel3 142 205 dur_cust_rel4 118 143 dur_cust_rel5 105 93 and in the model I do first wrote the new variables in the dataset: data test; set score; retain AVG_DELAY_L24M_cat DUR_CUST_REL_cat; AVG_DELAY_L24M_cat = put(AVG_DELAY_L24M,AVG_DELAY_L24Mgroup.); DUR_CUST_REL_cat = put(DUR_CUST_REL,DUR_CUST_RELgroup.); run; and the to the model: proc logistic data = test descending; class AVG_DELAY_L24M_cat DUR_CUST_REL_cat ; model bad_client= AVG_DELAY_L24M_cat DUR_CUST_REL_cat; run; And for all the other variables the frequency table is just like this two, I don't have any class without observation. I created the classes based on WOE and choose the variables for the model using the IV. I do took a look on the frequencies but I didn't saw anything suspicious. Could be something else? Thank you.

danirecchia · ‎08-09-2012

Thank you Reeza, I solved writing the new variable format on the dataset: data test; set scoring; retain AVG_DELAY_L24M_cat; AVG_DELAY_L24M_cat = put(AVG_DELAY_L24M,AVG_DELAY_L24Mgroup.); run;

danirecchia · ‎08-09-2012

Hello there, I am running a logistic regression with my original data and it is ok. But I also groupped them in classes, for example: age1: 18-25 years old age2 :25-30 years old age3: 30+ years old and now I wanto to run the modell with the groupped variables but when I do I receive the messagem about quasi-complete separation. The problem is, that I don't have a clue about which variable(s) are causing this problem. How should I proceed to fix it? Thanks!

Online Status	Offline
Date Last Visited	‎09-01-2015 07:11 AM

Re: report refresh - running sas code

report refresh - running sas code

Re: convert ddmonyy:hh:mm:ss.ss to mmyyyy

convert ddmonyy:hh:mm:ss.ss to mmyyyy

Re: how to compare two models

Re: how to compare two models

Re: how to compare two models

how to compare two models

Re: quasi-complete separation in logistic regression

Re: quasi-complete separation in logistic regression

Re: report refresh - running sas code

Re: report refresh - running sas code

Re: report refresh - running sas code

Re: report refresh - running sas code

report refresh - running sas code

Re: convert ddmonyy:hh:mm:ss.ss to mmyyyy

convert ddmonyy:hh:mm:ss.ss to mmyyyy

Re: how to compare two models

Re: how to compare two models

Re: how to compare two models

how to compare two models

Re: quasi-complete separation in logistic regression

Re: quasi-complete separation in logistic regression

Re: quasi-complete separation in logistic regression

Exclusion from a group from a class variable

Re: quasi-complete separation in logistic regression

Re: proc logistic with grouped variables

quasi-complete separation in logistic regression