**pbac**
(45d7e384)
*
at
30 Sep 09:10
*

Submitted 1.0.0

**pbac**
(2b10bc1f)
*
at
24 Sep 14:51
*

Ready for 1.0.0 submission

**pbac**
(2b10bc1f)
*
at
24 Sep 14:51
*

Ready for 1.0.0 submission

by making a "forecastmatrix" and "fmlist" classes such that e.g "bs.forecastmatrix", and "residuals.forecastmatrix" can be made and simlar. This will allow the use of bs() directly instead of bspline(), so actually nicer overall

**pbac**
(ce4b10c9)
*
at
07 Sep 11:48
*

misc

**pbac**
(34ed562d)
*
at
07 Sep 10:52
*

pi was not allowed in transformations in model(), now set as a spec...

**pbac**
(a0350cdb)
*
at
17 Aug 08:47
*

Fixed for check, submitted as version 0.10.0

**pbac**
(2478dbfe)
*
at
16 Aug 16:00
*

Added NAMESPACE and man, otherwise the install from the git is not ...

**pbac**
(f41468d5)
*
at
12 Aug 14:37
*

with kseqopt

**pbac**
(0541338a)
*
at
13 Jul 16:51
*

step_optim improved and misc other stuff

When the score is calculated in each step, make sure that only the same points are included for all models, or print warning if not.

Done, see comment in #8

In step_optim() it's now possible to give the fitting function, e.g. "fitfun=rls_fit" and the scores will be calculated, in each step separately, only using complete cases across horizons and models. Some extra information is printed and returned. So not across steps, that's not really possible.

In the fit functions the score() function could be used taking only complete cases across all horizons as default...

Optimize the parameters for each horizon, e.g. if kseq = c(1,6,18), then three optimizations are run, yielding the optimal parameters for each horizon. When fitting the model the other horizons, an interpolation could be done, e.g. for k=3 would be the average of the 1 and 6 horizons optimal parameters.

**pbac**
(08ae69e7)
*
at
02 Jul 22:08
*

made periodic function

We should then also consider a robust RLS https://scholar.google.dk/scholar?q=robust+recursive+least+squares&hl=en&as_sdt=0&as_vis=1&oi=scholart