Hi Sharon, I sent your note to sasoda@sas.com and copied you on it. Thank you! ... View more

Hi, Is this still an issue for you? You are correct that the individual file size limit is 1GB. Note that ODA also has a total storage limit of 5GB in your home directory plus 3GB in any courses you have created. You can check your storage limits by clicking on the second icon down on the right side of the ODA launch page:   If you are still having the issue, and it is not related to your total file storage, it would be helpful for you to post a screen shot of the error message in English or tell us its exact wording in English. ... View more

Hi, For distributions other than the lognormal distribution, you can use the ILINK option on the LSMEANS statement to get estimates on the original scale of the data. This will not change anything in your output when used with the lognormal distribution, though, since the lognormal distribution uses an identity link.  For the lognormal distribution- you have the manually back transform your data. This other communities post has relevant suggestions on how to do that. I have not tried this myself. Proc Glimmix data transform and backtransform - SAS Support Communities You can also try a different distribution, instead of the lognormal distribution, then use the ILINK option. ... View more

As others have mentioned, the MPRINT option does work on ODA. You can prove it to yourself by starting a fresh SAS Studio session (exit out of ODA, then come back in) and running a simple macro program. For example, you could run this program: options mprint; %macro mac1; proc print data=sashelp.cars(obs=10); run; %mend; %mac1; For your existing code- perhaps you have unbalanced quotation marks or your macro function did not correctly complete when being defined? In both those cases, SAS gets into a situation where any additional code is run as if inside a text string or inside a macro definition. There is no universal way to get out of those situations. Instead, you should exit out of ODA, restart SAS Studio, then fix your code. ... View more

Hi, PROC MIXED on ODA has the same functionality as the desktop version, including use of the OUTPM option. Please provide us your log and full MIXED output. That will help to diagnose the issue. ... View more

Hi, Thanks for the log and output. Let's call your two BY conditions 1A and 2B.  When you add the SOLUTION option on the RANDOM statement, I see in your output for 2B that the table "Covariance Parameter Estimates" stays the same, whereas the tables "Solutions for Fixed Effects" and "Estimates" change. It is puzzling that the estimates for the covariance parameters stay the same, but the estimates for fixed effects (which are functions of the covariance parms) change. Off hand, I don't know why adding the SOLUTION option on the RANDOM statement should change the model. Maybe the option changes the inverse matrix being used in estimates? As @Ksharp mentioned, you do have numerical problems with your model that you should fix.  Here is a blog with some more information about the note "Estimated G matrix is not positive definite": Convergence in mixed models: When the estimated G matrix is not positive definite - The DO Loop You do want to do something about a non-positive definite G matrix and don't want to continue with results where that is an issue.  Some things you could try: 1) Remove one of AVISITN or TIMEK from the RANDOM statement. This reduces your covariance matrix to make it simpler. I vote for removing AVISITN, since it has a zero variance estimate in your 1A case. 2) Specify the NOBOUND option on the PROC MIXED statement to allow variance components to be allowed to be negative. I don't have a lot of experience with doing this. 3) Try forming your covariance structure with a REPEATED statement instead of a RANDOM statement. For example, you could try a spatial power covariance structure with TYPE=SP(POW)(TIMEK). 4) Run this model in PROC GLIMMIX instead of PROC MIXED. You'd need to remove the METHOD=REML option, but otherwise, I think your code should run without edits. I doubt that using GLIMMIX would do anything about your non-positive G matrix, but it might help with the fact that your output is changing when the SOLUTION option is added. To really get to the bottom of this, I suggest you contact SAS tech support. Their help is included with your license: SAS Technical Support | SAS Support ... View more

Hello, It would be helpful if you could include your log and complete output. Thank you. ... View more

Awesome! I'm glad the response helped!  Regarding your comment about the FMTSEARCH option.. The st100d05.sas program saves the formats to a format library named work.formats, not stat1.formats. That is why your FMTSEARCH option code did not work. (Though- it's cool that you tried!) You can list all the formats in a particular library (and convince yourself that work.formats exists) by running this code below after you have run the st100d05.sas program. This prints all the formats in work.formats to the results window: proc format fmtlib library=work.formats; run; If you try running the code above, but you insert "stat1.formats" for "work.formats", the code throws an error, since stat1.formats does not exist. This is a bit too much information, but if you would like to save the STAT1 formats permanently, you can run the program st100d05.sas then run the program below. This copies the file "work.formats" to a file "formats.sas7bcat" that is located in the STAT1 Data folder. proc catalog catalog = work.formats; copy out = stat1.formats; run;   Then, instead of having to run st100d05.sas every time you open SAS, you can use this code below in order to access the stat1.formats format catalog. This is the same code you mentioned in your reply: options fmtsearch = (stat1); Good luck with the course! ... View more

@SteveDenham, The overall simulation results are discussed in this paper:  https://pubmed.ncbi.nlm.nih.gov/26089186/   Unfortunately, I don't show the results for trying out different optimization methods from GLIMMIX and MIXED. We just summarize the best one. For our set cases --which were low number of independent observations with high numbers of repeated measures-- the best method turned out to be TECHNIQUE=DBLDOG. To be honest, I have no idea why that one worked best! Thank you for asking! ... View more

Hi, While something else might be going on, it is worthy to note that the default optimization methods are not identical in MIXED and GLIMMIX. MIXED uses Newton-Raphson with ridging (and you cannot change this). You can change the optimization method in GLIMMIX using the NLOPTIONS statement with TECHNIQUE option. From the documentation I see this comment about the default optimization method in GLIMMIX: The default optimization techniques are TECHNIQUE=NONE for noniterative estimation, TECHNIQUE=NEWRAP for singly iterative methods in GLMs, TECHNIQUE=NRRIDG for pseudo-likelihood estimation with binary data, and TECHNIQUE=QUANEW for other mixed models. Personally, I had a simulation study I once ran that had widely different convergence rates depending on which method I specified on the NLOPTIONS statement. MIXED was also a comparator in my study. In the end, for my cases (which admittedly were unusual), I saw poor convergence rates for MIXED, the default in GLIMMIX, and many of the NLOPTIONS techniques, but did find one of the GLIMMIX techniques that had very high convergence rates. The moral is- convergence in GLIMMIX can change a lot based on what option you specify on the NLOPTIONS statement. ... View more

Hi, Are you using SAS studio through SAS OnDemand for Academics (ODA)? If so, below is some text from the ODA FAQ. I recommend you email ODA support at sasoda@sas.com and ask about the status of your account. Does my account expire? Access to SAS OnDemand for Academics can expire after an extended period of inactivity. However, to reactivate your access to SAS OnDemand for Academics, visit welcome.oda.sas.com enter your SAS Profile credentials and accept the terms of the license and the terms of use and conditions. Does my data or content expire? If your account is not reactivated after an extended period of time, SAS OnDemand for Academics will deactivate your account and archive your content. After an additional period of inactivity the archived content will be removed from the SAS OnDemand for Academics systems. Archived accounts are subject to having content removed without notice and this is at the discretion of the SAS OnDemand for Academics Team. Per the license agreement, SAS has the right to remove any Content at any time. ... View more

To add to the comments of @sbxkoenk, my understanding of your problem is that the default F test provided by PROC GLM is computing the same test computed by your code. For example, here is the output from one of the examples in the PROC GLM documentation:   The test you want is in the columns labeled "F Value" and "Pr>F" in the upper right of the first table. ... View more

Hi, Check out the documentation for the FOREST and HPFOREST procedures and make sure you have valid syntax for each of your statements. For example, you have this statement with the FOREST procedure: TARGET Default_Flag / LEVEL=BINARY; The only valid choices for the LEVEL= options are LEVEL=NOMINAL or LEVEL=INTERVAL. The error notes in the log can be hard to figure out, but this error is trying to tell you that LEVEL=BINARY isn't valid syntax: ERROR 22-322: Syntax error, expecting one of the following: INTERVAL, NOMINAL.     ... View more

Hello @JLF5806! The syntax of the both the SUBSTR and SUBSTRN functions is:   SUBSTR (string, start <,length>) In both cases, the function reads the string left to right, starting at the position specified by start and returning a string of the specified length. The two functions differ in how they handle zero or negative start and length values.   Your second example of SUBSTR("ABCDE", 1, 2) specifies a starting position of 1 and a length of 2, so the function starts at the letter "A" and returns two characters, resulting in the string "AB". Similarly, your first example of SUBSTRN("ABCDE", -1, 4) specifies a starting position of -1 and a length of 4. When the starting position is negative or zero, SUBSTRN returns a string that starts at the first position (here the letter "A") and adjusts the length of the string returned to account for the start value. If the starting position is 0, SAS subtracts 1 off the length. If the starting position is -1, SAS subtracts 2 off the length, etc. For a starting position of -1 and length of 4, think of it, for example, that you count -1, 0, 1, 2, but SAS only returns the characters starting at position 1, resulting in the string "AB". Check out Example 1- Manipulating Strings with the SUBSTRN Function- in the SUBSTRN documentation for another example. ... View more