Oh, yes, switch to MIXED or even (my favorite) GLIMMIX!   This example in the MIXED procedure may produce what you are looking for, if what you want is a random coefficients model https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_mixed_sect034.htm   Using that example as a template, and throwing in a plot, your code would look like proc sgplot data=example; reg x=X y=Y / group=ID; run; proc mixed data=example; class ID; model Y = X / solution; random Intercept X / type=un sub=ID solution; run; "type=un" specifies random intercepts, random slopes, and covariance between intercepts and slopes. You may find that a simpler structure (e.g., setting covariance to zero, or only random intercepts) might provide a better fit to your data.   The estimates of regression parameters (intercept and slope) for the population of IDs is produced by the SOLUTION option on the MODEL statement. ... View more

Is the MIXED procedure also telling you that the estimated G matrix is not positive definite?   Most likely, this result implies that there is little variation among reps for this response variable. The procedure has set the estimate to zero (hence, the SE is missing) and continued on its merry way. See Section III in this paper http://support.sas.com/resources/papers/proceedings12/332-2012.pdf   In the documentation https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_mixed_sect028.htm under Parameter Constraints, you'll find:   "For some data sets the final estimate of a parameter might equal one of its boundary constraints. This is usually not a cause for concern, but it might lead you to consider a different model. For instance, a variance component estimate can equal zero; in this case, you might want to drop the corresponding random effect from the model. However, be aware that changing the model in this fashion can affect degrees-of-freedom calculations.   If that doesn't seem to be the reason, then you'll want to post your code and an example dataset so that people here have more to work with.   ... View more

Thank you for clarifying your data set structure. I'm more sure now that we're on the same page.   Focusing on the covariate X: Because there are multiple (X,Y) observations (one for each level of Level2) within each level of Level3, we (or more properly, the statistical model) are able fit a regression of Y on X for each level of Level3; as you note, this set of regressions may have appreciable variance among intercepts, variance among slopes, and covariance between intercepts and slopes. These (co)variances are derived from the multiple Level3 regressions. Consequently, although you can assess whether there are random intercepts and random slopes, I'd say that assessment is "among" levels of Level3; there is no random intercept/slope among levels of Level2 because the model is using the different levels of Level2 (within each level of Level3) to define the regressions. I hope that make sense.   I failed to define "Xmean" and "Bmean". Xmean is the mean of the X values over the levels of Level2 for each level of Level3--it's like moving the X values up a tier, from Level2 to Level3, as if Xmean was measured at Level3. I hope that makes sense, too. This concept is addressed in the Singer paper (SES and MEANSES) I linked in an earlier response. Although I didn't intend them as centered variables, they certainly could be, and are in the Singer paper. If you center them correctly, both should be variable (i.e., not constant zero, although the mean would be zero). Should you center? Your call. If the model includes interactions (including polynomial terms, like X*X), then centering is very useful and potentially does reduce collinearity. In a model without interactions, it's less critical, I think. Centering doesn't hurt; you just have to rescale results to un-do centering if you want results on the original scales. Should you include Xmean and Bmean? Again, your call.   If it was me, because there are no covariates at Level1, I would compute the mean Y over the levels of Level1 for each level of Level2 within each level of Level3 and then use the mean Y as the response in the simpler, two-level model. Nothing wrong with an easier life  🙂 You would then be able to omit the second RANDOM statement. If the number of levels of Level1 are the same for all combinations of Level2 and Level3, then the statistical tests for fixed effects will be very similar, if not identical, to those from the three-level model. If the number of levels of Level1 varies dramatically among combinations of Level2 and Level3, then I might keep the three-level model.   I haven't looked in any detail at the paper you found with the macro for assessing assumptions. If you adequately understand how the macro is addressing assumptions, and know what the assumptions are and how to extract what you need from the MIXED procedure, you theoretically would be able to extend the methods to a three-level model. In a sense, your statistical model is a multiple regression in a mixed model, so you have all the assumptions associated with multiple regression plus the assumptions associated with a mixed model. A busy task, but not horribly difficult.   Good luck! ... View more