SCM

SCM Repository

[depmix] View of /todo
ViewVC logotype

View of /todo

Parent Directory Parent Directory | Revision Log Revision Log


Revision 99 - (download) (annotate)
Sun Mar 23 14:11:15 2008 UTC (11 years, 7 months ago) by ingmarvisser
File size: 2314 byte(s)
Final minor changes
TODO list for depmix package release 0.1-0

(priority to be determined)

- add help files for all functions:
	1) depmix.Rd: help on constructing models: done!!!!!
	2) depmix-class.Rd: detailed description of depmix class: Ingmar, done!!!!!
	3) depmix.fit.Rd: help on fitting models: Ingmar
		Other methods for depmix.fitted objects include:
		a) posterior: get posterior state sequence: we still need the code for this!
		b) logLik: done!!!
		c) AIC, BIC: done!!!
		d) getpars, setpars: done!!!!
		e) llratio: done!!!!!
		
	4) responses.Rd: detailed description on constructing response models: done!!!!!
	5) transInit.Rd: detailed description on constructing transition models: done!!!!!
	6) others: lystig, fb: done!!!!

- rework em function and add a working example for it: Maarten

- balance scale data example with covariate on initial probabilities
(this involves a latent class model rather than a hidden Markov model)
Ingmar: done!!!!!

- review summary/print functions for depmix and depmix.fitted: done!!!!!

- viterbi algorithm to get maximum a posteriori states for depmix model: Maarten, done !!!!!

- viterbi algorithm to get posterior probabilities as part of depmix.fitted: skip!!

- general documentation: depmix-introduction.pdf needs updating:
	1) check credentials
	2) add help for posterior
	3) 











TODO long term 

- generate initial values?!?!?

Examples

1) Discrimination learning example, need covariates with these data!

2) Factor depmix model as example of user specified model


Design issues

1) Data simulation according to specified model

2) Analytical gradients in lystig

3) Speed optimization: lystig in C?

4) Multi group possibilities: use group factor in call to depmix

5) Missing data options?

6) Standard errors of parameters


Other capabilities

1) Look at log and exp of matrices to have continuous time observations
(also see mlm)

2) Look at HMISC for mapply -> multivariate apply for mv densities

3) Look at package ks for mv normal mixture densities

4) Check flexclust package for an example of extending models

5) Check glmc pack for constrained glm models!!!!!

6) stochmod pack for stochastic models

7) use relevel function to recode factors into different base

root@r-forge.r-project.org
ViewVC Help
Powered by ViewVC 1.0.0  
Thanks to:
Vienna University of Economics and Business Powered By FusionForge