Version 4.4-0

1. gamlss

- i) the term.plot() did not plot (by mistake) a gam object. This is now corrected
- ii) the vis.lo() is introduced for plotting fitted loess curve
- iii) VC.test() for Vuong and Clarke test is introduced (use with caution)

iv) the function stepVGDAll.A() is introduced.

v) Q.stats() function is amended so the number of observations in each intervals add up to the total number of observations (that was a bug)

vi) the pcat() function for reducing the levels of a categorical

variable is introduced.

vii) the pbc() a more robust version of cy() is introduced.

2. gamlss.mx

- i) the function plotMP() is added

3. gamlss.data

- i) the Tokyo rainfall data is included in gamlss now as an example of cycle P-spline smoothing
- ii) the data set grip (hand grip strength) is added (for centile estimation)
- iii) the data set leukemia is added (for random effect analysis)
- v) the data oil is included

Dear GAMLSS Team,

thanks for constantly improving the great GAMLSS package!!

Since you`ve included the an interface function to use rpart() function within GAMLSS, I was wondering if you plan to also include an interface function to use the randomForest() function within GAMLSS. This would really improve the awareness of the GAMLSS package, also within the machine learning community.

Best,

Alex

Dear Alex

It would be nice to include randomForest() in GAMLSS.

To do so it requires prior weights.

I will try to look at the problem after Easter.

Thanks

Mikis

Dear Mikis,

thanks for taking up my request. For the implementation of Random Forests, you might want to also look into the “ranger” package, which is a fast implementation of random forests. It has a “case.weights” argument, which might be the weighting argument needed for your fitting algorithm to run.

Best,

Alex