Hi, there:
I run into some convergence issues when using proc model to estimate a highly nonlinear model. I tried to start from different starting point (using the option startiter) and raise the convergence criteria (for example, change the value from 0.000001 to 0.001), but nothing has worked. One hint I can get from the program is that the RPC value is always missing. Does it tell me something? In addition, I looked the correlation matrix of the explanatory variables. Although there are some eigenvalues are very small, but they are not too small.
Any input is highly appreciated.
Showing the code of the model you ran may help.
Also some of the output. It may be something like number of observations is insufficient or something else data related that we really would be guessing in the dark at this point.
Hi, there:
The model is fairly complicated. There are about 1 million observations and I checked the variance covariance matrix of the explanatories is full rank. However, the computation of RPC is missing, but everything else can show. That makes me wonder what it tells me.
Don't miss out on SAS Innovate - Register now for the FREE Livestream!
Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.