|
|
Thresholds fixed at +/-15 |
|
Message/Author |
|
Daniel posted on Wednesday, September 17, 2003 - 6:20 am
|
|
|
I received the following message in my output. However, none of the thresholds were -15 or 15 in the final estimates. Can I trust my results? IN THE OPTIMIZATION, ONE OR MORE LOGIT THRESHOLDS APPROACHED AND WERE SET AT THE EXTREME VALUES. EXTREME VALUES ARE -15.000 AND 15.000. THE FOLLOWING THRESHOLDS WERE SET AT THESE VALUES: * THRESHOLD 1 OF CLASS INDICATOR SMOKE12 FOR CLASS C#1 AT ITERATION 10 * THRESHOLD 2 OF CLASS INDICATOR SMOKE12 FOR CLASS C#1 AT ITERATION 10 * THRESHOLD 1 OF CLASS INDICATOR SMOKE12 FOR CLASS C#2 AT ITERATION 2 |
|
|
Please send your output to support@statmodel.com. |
|
Jessica posted on Monday, April 03, 2006 - 7:25 am
|
|
|
Hello. I am running a latent class (2, 3, and 4 classes) analysis with 6 binary indicators. The 2 class analysis ran fine, but for the 3 and 4 class analyses, I received the same message as the previous person. For the 3 class analysis, one threshold 1 error was reported, and for the 4 class analysis, two threshold 1 errors were reported. In the output, these probabilities are fixed to either 1 or 0. Is there anything I can do to "fix" this problem? Thank you. |
|
|
Having thresholds fixed at -15 or +15 is not a problem. It just means that for endorsing that particular item the probability is one or zero in that class. It can be helpful for defining the class. |
|
Jessica posted on Monday, April 03, 2006 - 7:45 am
|
|
|
Thank you. I was just concerned that this was indicating that my model was not identified (although both analyses have fine condition values, standard errors, etc.). |
|
Back to top |
|
|