İ am working on a binominal regression analysis for a retrospective study. Outcome has two option. Positive or negative. İn my Best regression model that i have found there are a continuous and an ordinal variable (1,2,3...10 ). And İ have used Jamovi and Spss for analysis and the model is fit. The problem is my spss subcription is almost out of date so a few weeks later jamovi Will be my only option for my research. İ Wonder is there any option or alternative test for hosmer le.. Test in jamovi, jasp, or another free statistical software?
Sure try R. Attached is an example. Using this method (adaptive lasso) you get an optimal solution from your full model variables not "best" predictive model because this method satisfies an oracle property-see paper and references. To do this, just follow the protocol in the paper. I can send you the R programs you need or you can download them from D. Boos" website-see paper references. Contact me if you have questions. Best, D. Booth
Jamovi is using R under the hood. If you like using Jamovi (rather then R directly or via RStudio), have a look at the Rj Editor (https://blog.jamovi.org/2018/07/30/rj.html) to use David's R code directly within Jamovi.
Not sure how well this works (haven't tried it myself), but from the description I read that this should not be much of a problem.
Have a look at this discussion
https://stats.stackexchange.com/questions/169438/evaluating-logistic-regression-and-interpretation-of-hosmer-lemeshow-goodness-of/207512#207512
It says that that test you seek is no longer recommended! And points you to others and a model-based approach , all implemented in R.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software