Remember!  With respect to the SAS Code Node in SAS Model Studio : Modification of the data that you want to pass on to subsequent nodes and to publishing must be done within the scoring code pane.   See: The SAS Code Node in SAS Model Studio versus SAS Enterprise Miner https://communities.sas.com/t5/SAS-Communities-Library/The-SAS-Code-Node-in-SAS-Model-Studio-versus-SAS-Enterprise/ta-p/813958 Last update: 17-MAY-2022 Updated by: BethEbersole   Why not create new variables in advance? In SAS Studio, for example. Then these will already be in the CAS table that serves as the Analytical Base Table (ABT).   Ciao, Koen ... View more

SAS Model Studio does not allow for free-form drawing of arrows / connections like SAS Enterprise Miner does.   The core difference is that Enterprise Miner's visual process flow allows for more manual customization, including the free drawing of connections, while Model Studio focuses on a more streamlined and less manual approach to data pipeline creation by auto-generating connections between distinct visual components.   See here for more info : Model Studio for SAS Enterprise Miner Users: Part 5, Building Models…Let’s get physical! ( by JThompson ) Ciao, Koen ... View more

@Jasmijn  You start with a well-chosen team name. That's promising. Ciao, Koen ... View more

@Serinac01 wrote: I used a simple decision tree and downloading that score code still resulted in astore procedure Yes , a simple decision tree also has its score code in an astore binary object.  As confirmed by the table on this page : SAS Help Center: Scoring Your Models Linear, logistic and quantile regression (a.o.) produce data step score code.   You can make your decision tree with PROC HPSPLIT. (the HP-prefix is for High-Performance , it's not a CAS-procedure but it works multi-threaded) PROC HPSPLIT  produces a file that contains SAS DATA step code for scoring new data produces a file that contains node rules You can integrate that PROC HPSPLIT in your Model Studio pipeline (SAS code node).   Ciao, Koen ... View more

Hello , What's your destination platform / destination environment (where everything is deployed in production)? Is your production environment still running SAS 9 ? Ciao, Koen ... View more

Hello ,   There are two types of score code that Model Studio nodes can create: DATA step or analytic store. To generate score code for an entire pipeline, the score code for each node producing score code is appended together into a single DATA step. When the nodes in a pipeline produce multiple analytic stores, or one or more analytic stores and DATA step score code, an EP score code file is created. EP score code represents the score code produced by these pipelines. (the above is from : SAS Help Center: Scoring Your Models ... there is also a table that tells you which node produces what.) You can run this code outside Model Studio. For complex models, modeling logic may not be stored in a human-readable equation or series of rules but, rather, a machine-readable file. Common file type that may represent models include Pickle for Python models; Analytical store (ASTORE) for SAS models; and rdata, rds, or rda for R models. When the model is saved as an object it must be loaded before it can be used to score data. (the above is from : Training code, scoring code, and what makes a model - The SAS Data Science Blog)   I guess you are looking for simple data step score code (without analytic stores) that you can take anywhere ... but you can't get full data step score code for complex models. Ciao, Koen ... View more

Note that you can also use PROC MIXED and PROC GLIMMIX for fitting a model to panel data. Due to different (estimation) methods used , the results will not be identical to those from PROC PANEL. But PROC MIXED and PROC GLIMMIX do have an ESTIMATE statement.   See this (dated) article for a comparison between PROC PANEL and PROC MIXED. http://www2.sas.com/proceedings/forum2007/170-2007.pdf Usage Note 22114: Analyzing unbalanced panel data using PROC PANEL or PROC TSCSREG https://support.sas.com/kb/22/114.html (PROC MIXED is not the primary focus, but it is covered) Usage Note 22115: Estimating a model using the Parks method with unbalanced panel data https://support.sas.com/kb/22/115.html   The old TSCSREG procedure is also for panel data , but it does not have an ESTIMATE statement either.   Ciao, Koen ... View more

See SAS Macros for Experimental Design and Choice Modeling   And maybe @WarrenKuhfeld reads along.   Ciao, Koen ... View more

@sasalex2024 wrote:   That said, I’d appreciate any thoughts on the UCM and CUSUM issues I mentioned above. Thanks again. This is how I would use the CUSUM method for Structural Break Detection / Change Point Detection in time series: Make an appropriate time-series model for your time series (it should fit the data well). Use PROC ARIMA or PROC SSM (for linear state space models) or PROC UCM or ... then proceed to Model-based Break detection methods that use CUSUM and/or CUSUMSQ statistics. CUSUM and CUSUMSQ statistics are computed using the cumulative sum and cumulative sum of squares of "one-step-ahead residuals" , respectively. With only a little bit of effort , you can use these residuals to do the CUSUM/CUSUMSQ-based change point detection as a post-fit diagnostic step. The CUSUM/CUSUMSQ-based change point diagnostics can be complementary to the De Jong-Penzer algorithm-based change point diagnostics. In particular, the CUSUM/CUSUMSQ-based change point diagnostics can be useful to detect changes in the variance of the process. Ciao, Koen ... View more

If you have longitudinal data without too much autocorrelation (serial correlation) , you can look here : Usage Note 70498: Interrupted time series analysis: Assess the effects of an intervention on subjects observed longitudinally https://support.sas.com/kb/70/498.html Usage Note 61830: Estimating the difference in differences of means https://support.sas.com/kb/61/830.html Otherwise (in case of time series with serious autocorrelation ) , you can look here : Intervention Models and Interrupted Time Series (PROC ARIMA) https://go.documentation.sas.com/doc/en/etsug/15.3/etsug_arima_gettingstarted31.htm Example 8.6 Detection of Level Changes in the Nile River Data (PROC ARIMA) https://go.documentation.sas.com/doc/en/etsug/15.3/etsug_arima_examples06.htm I don't want to detract from the PROC UCM checkbreak method, which is also very good. PROC AUTOREG is another alternative. Ciao, Koen ... View more

Hi,   I only have 2 minutes right now ... so my quick suggestion is to use the missing value trick for scoring a regression model. The missing value trick for scoring a regression model - The DO Loop   The trick relies on two features of SAS software: The first data set contains variables X and Y. The second contains only X. If the two data sets are concatenated, a missing value is assigned to Y for each value of X in the second data set. When a regression procedure encounters an observation that has a missing value for the response variable, it does not use that observation to fit the model. However, provided that all of the explanatory variables are non-missing, the procedure does compute the predicted value for that observation. Good luck, Koen ... View more

Hello ,   Measuring Prediction Error in PROC HPFOREST The HPFOREST procedure computes the average square error measure of prediction error. For a binary or nominal target, PROC HPFOREST also computes the misclassification rate and the log-loss. See SAS Help Center: Measuring Prediction Error Oob refers to  Out-Of-Bag Estimates. SAS Help Center: Example 7.1 Out-Of-Bag Estimate of Misclassification Rate Ciao,  Koen ... View more

You need an SMTP e-mail host ! How to use Gmail to send a message from a SAS program - The SAS Dummy   I think you need  type='text/html' instead of plain text ! Use SAS to send an email that embeds a graph in the body of the email - SAS Users   Ciao, Koen ... View more

Another customer was facing the same problem and referred to this topic in his query to me. In the end, it turned out it was not due to a (lack of) memory problem there ...but due to the fact that he had used an integer variable as a by-variable. That can't be done. The by-variables in the forecasting hierarchy should all be character variables (varchar e.g.).   Ciao, Koen ... View more