Comparison of sparseS() in fullPath and redesign-code

aalfons / robustHD

Robust methods for high-dimensional data, in particular linear model selection techniques based on least angle regression and sparse regression.

GNU General Public License v3.0

10 stars 6 forks source link

Comparison of sparseS() in fullPath and redesign-code #30

Closed voellerer closed 2 years ago

voellerer commented 9 years ago

Problems in redesign:

1) number of iterations in M-scale changed inside the loop. Thus, after first sample, all fixed point iterations are done until convergence

2) After initial I-steps continueSteps needs to be set to TRUE. (Now it often happens that no further I-steps are carried out)

3) in sparseS, the objective function can also increase. Thus the difference between the previous and the current criterium of an I-step can be negative. Therefore, an absolute value needs to be applied.

aalfons commented 2 years ago

sparseS() will no longer be developed further. Users can use package pense instead.