I did move this topic to 'Statistical Procedures' - board. And calling @clayt85 .   Koen ... View more

I don't understand ... why would a data step behave differently when being part of a macro? The '_data_' works identically outside and within a macro, no?   Cheers, Koen ... View more

Okay ... a bit of context here ... for the readers of your post.   This is about a Discrete Choice Experiment and you are using the (quite old) SAS macro's explained on this page: SAS Macros for Experimental Design and Choice Modeling   Here are some good resources (also discussing topics like number and size of choice sets): mr2010f.pdf mr2010c.pdf Maybe @WarrenKuhfeld can help further?   Ciao, Koen ... View more

Okay, probably PRINTMISS option in PROC TABULATE behaves like the SPARSE option in PROC FREQ. The PRINTMISS / SPARSE option(s) provide "all possible combinations of levels of the variables in the input-table, even when some combination levels do not occur in the data". The PRINTMISS / SPARSE option(s) cannot know that one class variable is hierarchically nested in another one.   Can't you enrich your input dataset with state / region / county records that are absent in your current input dataset? Just put missing values for all other variables. No need to do that "manually" , just find yourself a geo-map dataset (state / region / county) and execute a merge.   Koen ... View more

FIRTH option in MODEL statement   FIRTH  = Firth’s bias-reducing penalized likelihood method ... available for binary response models only !   FIRTH option in MODEL statement performs Firth’s penalized maximum likelihood estimation to reduce bias in the parameter estimates (Heinze and Schemper 2002; Firth 1993). This method is useful in cases of separability, as often occurs when the event is rare, and is an alternative to performing an exact logistic regression. See the section Firth’s Bias-Reducing Penalized Likelihood for more information. ( Note: The intercept-only log likelihood is modified by using the full-model Hessian, computed with the slope parameters equal to zero. When fitting a model and scoring a data set in the same PROC LOGISTIC step, the model is fit using Firth’s penalty for parameter estimation purposes, but the penalty is not applied to the scored log likelihood. )   Heinze, G., and Schemper, M. (2002). “A Solution to the Problem of Separation in Logistic Regression.” Statistics in Medicine 21:2409–2419. Firth, D. (1993). “Bias Reduction of Maximum Likelihood Estimates.” Biometrika 80:27–38.   SAS Help Center: Firth’s Penalized Likelihood Compared with Other Approaches   This FIRTH method has the advantage of being much faster and less memory intensive than exact (Exact Conditional Logistic Regression) algorithms, but it might not converge to a solution.   Ciao, Koen ... View more

TABULATE procedure -- TABLE statement :   PRINTMISS prints all values that occur for a class variable each time headings for that variable are printed, even if there are no data for some of the cells that these headings create. Consequently, PRINTMISS creates row and column headings that are the same for all logical pages of the table, within a single BY group. MISSTEXT='text' supplies up to 256 characters of text to be printed for table cells that contain missing values. MISSTEXT={<label='text'> <STYLE=style-override(s)>} supplies up to 256 characters of text to be printed and specifies a style override for table cells that contain missing values. For details about the arguments of the STYLE= option and how it is used, see  STYLE= in the TABLE statement.   table ... ... / printmiss misstext='0'; Ciao, Koen ... View more

@hr667 wrote: This is close to what I want, but I would like to remove the '0' rows which represent missing. I tried setting these values to missing, but the table would then only include those who were consented i.e. remove observations from the table that had missing values and make the screened value incorrect.   I also would like to add formatting to the box directly to the right of the variable labels. It is currently outputting shaded blue, and I would like to remove that to match the formatting on the rest of the table. 1.)  To remove the '0' rows : To the TABLE statement and after the forward slash ( / ) you add : printmiss MISSTEXT='0' ... or just ... printmiss To the PROC TABULATE statement you add : out=tabulardata. In the tabulardata dataset (to be found in WORK library) you can delete (programmatically with a WHERE statement) the rows you do not want to display and then run a 2nd PROC TABULATE. The 2nd PROC TABULATE is only displaying ... not summarizing / not aggregating / not calculating! Make sure the screened (non-zero) values stay identical to the ones in the original table. Check this diligently. 2.)  To get rid of the shaded blue next to the variable label : Proc TABULATE does NOT have a table template like (for example) PROC FREQ and PROC MEANS. But the 'look' of PROC TABULATE output is impacted by the STYLE= overrides in the procedure syntax. And you can change the style by means of a style template. ODS HTML, proc tabulate, proc template, how to change style? https://communities.sas.com/t5/ODS-and-Base-Reporting/ODS-HTML-proc-tabulate-proc-template-how-to-change-style/td-p/160497 PROC TABULATE – Building Tables with Style Michael Eberhart, MPH, Department of Public Health, Philadelphia, PA https://www.lexjansen.com/nesug/nesug10/hw/hw04.pdf Ciao, Koen ... View more

How to choose a seed for generating random numbers in SAS? By Rick Wicklin on The DO Loop June 1, 2017 https://blogs.sas.com/content/iml/2017/06/01/choose-seed-random-number.html   It says ( amongst others ) A seed value specifies a particular stream from a set of possible random number streams. When you specify a seed, SAS generates the same set of pseudorandom numbers every time you run the program. However, there is no intrinsic reason to prefer one stream over another. The stream for seed=12345 is just as random as the stream for the nine-digit prime number 937162211.   BR, Koen ... View more

@Ronein wrote: 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? Gini in-sample (training data) = 80% Gini out-of-sample and out-of-time = 82% The Gini coefficient measures separation power. Similar to how it is used in economics to measure inequality. However for credit risk, the higher the Gini, the better.   So, I don't see the problem ... unless you fear a "too good to be true"-type of error.   Ciao, Koen ... View more

Sorry for still another link ... but I would do it this way : Home > Analytics > SAS/IML> find roots of function https://communities.sas.com/t5/SAS-IML-Software-and-Matrix/find-roots-of-function/td-p/794659   A quadratic function (equation) may have one, two, or zero roots. If the discriminant is less than 0, then the quadratic has no real roots. (but the quadratic equation still has two complex-valued roots) If the discriminant is equal to zero then the quadratic has equal roots. (two roots with the same value, called a double root) If the discriminant is more than zero then it has 2 distinct roots. Ciao, Koen ... View more

Find the root of a function by using the SAS DATA step By Rick Wicklin on The DO Loop August 19, 2020 https://blogs.sas.com/content/iml/2020/08/19/find-root-function-sas-data-step.html A simple way to find the root of a function of one variable By Rick Wicklin on The DO Loop February 5, 2014 https://blogs.sas.com/content/iml/2014/02/05/find-the-root-of-a-function.html Finding the root of a univariate function By Rick Wicklin on The DO Loop August 3, 2011 https://blogs.sas.com/content/iml/2011/08/03/finding-the-root-of-a-univariate-function.html If you use SAS Viya and SAS Studio 5.2 or later AND your site licenses SAS/IML , then there's a code snippet called: Find Roots of Nonlinear Equation : Enables you to find the roots of a function of one variable. Finding the root (or zero) of a function enables you to solve nonlinear equations. Paper 3099-2019 Find a Function Root By Genetic Algorithm Xia Ke Shan, iFRE Inc., Beijing, China Kurt Bremser, Allianz Technology GmbH, Austria (Presenter) https://support.sas.com/resources/papers/proceedings19/3099-2019.pdf root finding, proc optmodel https://communities.sas.com/t5/SAS-Procedures/root-finding-proc-optmodel/m-p/298259 Halley's method for finding roots  By Rick Wicklin on The DO Loop August 24, 2016 https://blogs.sas.com/content/iml/2016/08/24/halleys-method-finding-roots.html Ciao, Koen ... View more

>> ... , I've replied to my own question with a code, do you think it is acceptable as a solution? Thank you very much again.   It's oké. But you need to terminate PROC ARIMA with a QUIT; statement. PROC ARIMA is an interactive procedure. Because the PROC ARIMA is followed by a data step (in your code) , there's an implicit QUIT ... but still I prefer to put it. If you only submit the PROC ARIMA part , you will notice the last statement (the forecast statement) is not executed. The output dataset with residuals is not created.   Interactive procedures in SAS - The DO Loop   Running interactive procedures in SAS Studio and SAS University Edition - The DO Loop   Do you write unnecessary SAS statements? - The DO Loop   Ciao, Koen ... View more

SAS/ETS User's Guide -- AUTOREG Procedure Testing for Heteroscedasticity SAS Help Center: Testing for Heteroscedasticity SAS/ETS User's Guide -- MODEL Procedure Heteroscedasticity SAS Help Center: Heteroscedasticity BR, Koen ... View more