Hello, I have one straight forward question and one a bit more open regarding Model Studio (MS). I was wondering if anyone could help me out with grasping the main idea behind MS.   1. Is there a possibility to see what code exactly is invoked after using a Native Node in MS for validation purposes? 2. I was thinking about general idea behind Model Studio. As I see it, it is possible to click out a simple model by using Native Nodes which change &dm_data which represents the input data and its metadata automatically. I was wondering what if I am not satisfied with Native Node options (or if answer to the first question is negative) and whether it really makes sense to use MS just for its integration with Model Manager and handling scoring code? The downside as I see is that there are a lot of processes being invoked behind the scenes within every Node. All built-in macros have a purpose of handling data and metadata of the project and therefore technical tables are created to support the ongoing process in the pipeline. That also means that for instance just to invoke macro parameters from a parent node, table has to be created with macro parameters and its values, register using built-in macro and then fetched from a child node as on example shown on github. Also, looks like MS is designed in a way that process single input table and handles it and its metadata across the node. However, when complex computations and table handling is needed, it becomes a bit inconvenient, but does not necessarily have to be a bad thing. It imposes a new approach for modelers to build SAS code rather then handling one table into many in variety of ways while looping and working within WORK library. Seems like nodes impose a different approach to handling input data. To sum it up, I am wondering what are your thoughts or whether anyone else had troubles with deciding whether to stick with SAS Studio while migrating their processes to Viya or chose Model Studio for modeling purposes without using Native Nodes, even if that means some troubles with migrating current SAS 9.4 code to Code Nodes in Model Studio.   I would love to hear about your experiences with Model Studio and your thoughts about it.   Kind Regards ... View more

Hey Guys, I was wondering if you could help me out with this problem. So I have two tables named Country_1 and Country_2 and they have variables as in the example below: data WORK.Country_1; infile datalines dlm='~'; input Date_To:ddmmyy10. Weight_de:8. pct_GDP_DE:8. pct_UR_DE:8.; format Date_To:ddmmyy10.; datalines; 31/03/2004~0.4~0.586206896551724~0.37037037037037 30/06/2004~0.4~0.568965517241379~0.314814814814815 30/09/2004~0.4~0.517241379310345~0.259259259259259 31/12/2004~0.4~0.46551724137931~0.185185185185185 31/03/2005~0.4~0.258620689655172~0.12962962962963 30/06/2005~0.4~0.172413793103448~ ;;;; data WORK.Country_2; infile datalines dlm='~'; input Date_To:ddmmyy10. Weight_nl:8. pct_GDP_NL:8. pct_UR_NL:8.; format Date_To:ddmmyy10.; datalines; 31/03/2004~0.6~0.775862068965517~0.685185185185185 30/06/2004~0.6~0.793103448275862~0.648148148148148 30/09/2004~0.6~0.758620689655172~0.592592592592593 31/12/2004~0.6~0.741379310344828~0.259259259259259 31/03/2005~0.6~0.448275862068966~0.203703703703704 30/06/2005~0.6~~0.166666666666667 ;;;; The problem is to multiply dynamically third column from first DS by its weight and add it to the third column from second DS multiplied by its weight. In this case they are two columns, but I do not know how many columns are in those datasets (can be more). Nevertheless, both DS have the same number of columns. First two are always DATE_TO and Weight and the rest of the columns are positioned by its name, so the multiplication can be done either by their position or part of its name which is common.    So that would be: Weight * pct_GDP_DE + Weight * pct_GDP_NL = Col1 Weight * pct_UR_DE + Weight * pct_UR_NL = Col2 I checked two possibilites: arrays and proc iml, but can not really start to do it dynamically.   Thank you for your time and possible solutions, m ... View more