EJ
About
- Username
- EJ
- Joined
- Visits
- 2,557
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Correct. The new version should be available this week. E.J.
-
Hi Mateus, There isn't, as far as I know. But most people find it relatively straightforward to interpret posterior probability. Actually, our "pizza plot" method works with posterior probabilities. See https://www.bayesianspectacles.org/l…
-
Hi Emma, Not at the moment in JASP, although perhaps you can with Johnny's R code -- I'll ask him. E.J.
-
Your conclusion for Table 2 should be the other way around. 0.106 under the null is 1/0.106 = 9.43 under the alternative. That this is support *against H0* and *in favor of H1* is also evident from the P(M|data) column. E.J.
-
Hi Emma, For the regular t-test, no MCMC is needed. In fact we try to avoid MCMC whenever we can. Not all Bayesian inference uses MCMC! However, for Kendall's tau we do use a sampling-based methodology and you are right, we don't give Rhat, and I th…
-
Hi Branden, At the moment you cannot change variable rows into columns using JASP. You can make this a feature request on our GitHub page (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/) Cheers, E.J.
-
JASP does not currently offer this. You could issue a feature request through our GitHub page (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/) Cheers, E.J.
-
Hi Mateus, This happens because "compare to best model" is the default setting, and the first row gives the model of reference. The BF in the first row is always 1 because the model is compared to itself. In the first table there is a smid…
-
Hi Sameha, No, that is not valid. These are factors and need to be treated as such. Cheers, E.J.
-
Hi Justin, There is no work-around yet, AFAIK. Cheers, E.J.
-
:-) Well the 0.15 release took a lot of effort, with lots and lots of work hidden from view, meaning it is key to check everything really carefully.
-
We are in the last stages of checking the new version
-
The team member who is ideally suited to implement this has had his hands full, but I am occasionally reminding him that this would be a great feature to add! Cheers, E.J.
-
This depends on what kind of Bayesian you are. A subjective Bayesian will have a prior plausibility assignment for every possible hypothesis. Data come in, the plausibilities are updated, and that's it. No adjustment needed. An objective Bayesian w…
-
Can you be a little more precise? Maybe a concrete example will help, with links to existing methodology (say in R) Cheers, E.J.
-
Hi Guan, Could you show a screenshot? Cheers, E.J.
-
As you can see from the code, the error is thrown by lavaan. We can look at it, but ultimately we might refer you to the lavaan community. We'll have to see. Cheers, E.J.
-
Yes, no option now, but I assume a graphics editor will do the job Cheers, E.J.
-
I'm not sure, but I think the first question is whether you'd expect the same result without the mediator, and I don't think you do. See for instance https://stats.stackexchange.com/questions/22718/what-is-the-difference-between-linear-regression-on…
-
In discrete data models, the log likelihood will always be negative (b/c the data is associated with probabilities). In continuous data models, the log likelihood is about the density, and that may be lower than 1 as well. Likelihood is defined up t…
-
Hi Martyna Yes, you can report that. Or you can rerun the procedure several times and get an idea of the % that way. And thanks for your kind words! Cheers, E.J.
-
Hi Ming, The method that JASP uses depends on the specific analysis problem. We try to use analytic solutions whenever we can, or integrate out variables before switching to a numerical procedure. Most often we integrate over a single parameter, but…
-
I agree it is wrong in the sense that there is nothing in between the discrete categories. However, the lines often make it easier to see the pattern (perhaps not in this case). Ideally the lines would be optional. I'll make a note of this. Cheers, …
-
I've attended the relevant team member to this. E.J.
-
I'll ask our expert. E.J.
-
This is strange. In fact, it should not happen, because a 95% CI contains all the values that would not be rejected using an alpha=5% test. The conflict could be due to the noise in the bootstrap...can you provide a concrete example so we can reprod…
-
I'll ask our expert E.J.
-
Hmm we'll provide more descriptive graphs in the upcoming release, but not this one, and I agree it it useful. I'll make a note of this. You could also make it a GitHub issue (https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/) Cheer…
-
Hmm this is really strange. It would be great if you could make an issue of this on our GitHub page. See https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/. You are using JASP 0.14.1? Cheers, E.J.
-
Hi Myrthe Yes, we generally also find that the Bayesian results are more in line with the frequentist results from lme4. In the preprint that Johnny mentioned you'll see a discussion of adding those random slopes. I agree with you that these models …