Commit Graph

103 Commits

Author SHA1 Message Date
Brian Albert Monroe 4cda6aec76 ML.winner: Check if model is in the ml.dataframe
Don't use models that weren't in the dataframe to pick winners
2017-05-12 15:24:50 +02:00
Brian Albert Monroe 68bca726d5 Add ability to select models in ML.all
Similar to ML.winner, pass character vector of model names to estimate
2017-05-12 15:07:00 +02:00
Brian Albert Monroe db78f1a1e6 Add option to choose which winners are considered
So for instance, we can figure out which model is the winner among just
EUT and Prelec.
2017-05-12 11:29:59 +02:00
Brian Albert Monroe 7863c35d3e Patch bump
This patch should have been 0.1, but I forgot to reset when bumping the
minor version number
2017-03-07 12:09:21 +02:00
Brian Albert Monroe 29ac4bd9eb RcppExports has changed, don't know why... 2017-03-07 12:08:14 +02:00
Brian Albert Monroe 34069530d0 Update ML.winner to be much, much faster 2017-03-07 12:07:54 +02:00
Brian Albert Monroe a74b4327a2 Bump the minor version number 2017-03-04 14:49:12 +02:00
Brian Albert Monroe e29e13b27a Add ML.winner function to redeclare winners
Also update ML.all to allow for different optimizers for different
models
2017-03-04 14:47:52 +02:00
Brian Albert Monroe ab24e48686 Add option to used pooled estimates as initial values in ML.all 2017-03-03 17:19:14 +02:00
Brian Albert Monroe 42f59b2d66 Fix ML.df winner picker code
It handled winner incorrectly if EUT didn't converge as well as not
handling 0 model convergences correctly
2017-02-28 12:14:37 +02:00
Brian Albert Monroe 63417f1ced Add calculation of winners to ML.dataframe
It seems more proper to do this kind of thing here as I want to keep
ML.all to just be a list of solved models.
2017-02-10 16:14:19 +02:00
Brian Albert Monroe 907254e169 Roxygen Update changed some documents 2017-02-10 16:09:22 +02:00
Brian Albert Monroe edf0e45015 Fix typo from last commit, add pfunc to results 2017-02-02 22:00:45 +02:00
Brian Albert Monroe e174c809a4 Add par to non-split in ML.all 2017-02-02 21:42:13 +02:00
Brian Albert Monroe 5292469336 DEoptim moved to suggests, I never use it.
I might eventually think that maxLik or even optimx might deserve to be
a suggests as well as only one or the other is really needed, but I use
them both all the time, so probably not.
2017-02-02 19:54:56 +02:00
Brian Albert Monroe c6a5a56a76 Last commit should have had a bugfix version bump 2017-02-02 19:42:12 +02:00
Brian Albert Monroe 9874533c94 Small spacing issues, change default value for ML.all 2017-02-02 19:40:56 +02:00
Brian Albert Monroe 594b6ea1d0 Add in ML.all to estimate all models
ML.all estimates all the different pfuncs available on a full dataset or
one split by a factor in the dataset. ufuncs will still
have to be passed as an argument, as well as with/without context. This
function brings in one of my other packages "ctools" to implement
parallel estimation of the split dataset. If you're not using ctools,
check it out, its super useful for these kind of situations.
2017-02-02 19:31:16 +02:00
Brian Albert Monroe 2a25bec56c Add in function ML.dataframe put main data on one row of dataset
This function takes a list which contains a list of estimated models.
The main use of this function will be for a yet written, ML.all
function which will split a dataset by a factor and estimate all models
on the split data. This ML.dataframe function will allow for all this
data to be easily condenced into a data.frame object. Many users might
find this a useful way to share the results of estimates with users of
Stata and other statistics softwares.
2017-02-02 17:17:48 +02:00
Brian Albert Monroe b42c585e5f Minor version bump for Contextual/Fechner change in ML 2017-02-01 16:39:50 +02:00
Brian Albert Monroe 5e94f8a800 Add in flag to use Fechner instead of Contextual for ML
Some people don't use contextual utility. I suppose at some point I'll
incorporate more exotic stochastic functions, and probably utility
functions as well.
2017-02-01 16:38:18 +02:00
Brian Albert Monroe dd8eafe8ca Add custom convergence code to ML output 2017-01-26 15:28:49 +02:00
Brian Albert Monroe 996da26665 Allow ML to pass testpars, add eut.pval to main options
The recovery on the EUT pval thing only works for non-mixture models
2017-01-24 22:30:40 +02:00
Brian Albert Monroe edf3f6638e Merge kiwi:thesis/MSL 2017-01-24 00:58:05 +02:00
Brian Albert Monroe 443eb4f77a Bug fix version bump 2017-01-24 00:49:34 +02:00
Brian Albert Monroe c91fc0adfd Just, so many bugs were needed squashing in MSL
Noticed that some insturments were bombing out with errors saying that
the function couldn't even evaluate once. Did some investigating, found
that uvec indicies were for some reason not grabbing the instrument
unless thew were transposed, but also the choice vectors stuff wasn't
pulling in the right choice data. Not by a long shot. Found that this
could be corrected by actually using the c_index vector that had been
defined but unused, I don't know why it wasn't. So much, Do not use the
immediately previous commit.

If you want to revert behind this commit, use 966dfaf if you need MSL
to estimate properly. Note this won't give you the non-linear transforms
from nlWaldTest.
2017-01-24 00:49:02 +02:00
Brian Albert Monroe a9da0499ab Update description bugfix version bump 2017-01-23 00:00:55 +02:00
Brian Albert Monroe 1305f6d702 MSL functions weren't handling covariate addition correctly
The basepars were essentially being replaced by the last dem * the
last dempar. This totally erased the effect of the main par. This was
replace so dem*dempar is added to basepar.

Also kkt2 was decreased to make it harder to converge.
2017-01-22 23:59:24 +02:00
Brian Albert Monroe 5c8b881840 Update package version for bugfix 2017-01-22 22:08:04 +02:00
Brian Albert Monroe 596d87f8cc Minor update. Allow NM for MSL add parnames to MSL 2017-01-22 22:07:07 +02:00
Brian Albert Monroe 4c56b06914 Mostly spacing issues, but some parsing differences 2017-01-22 16:07:57 +02:00
Brian Albert Monroe a2f0582c83 Add nlWaldTest package for conf intervals and SEs
These functions are all non-linear, but I've been using linerar
transformations to recover standard errors from the hessian. Woops...
The nlWaldTest package provides a way of retrieving the non-linear
confidence intervals, from which the SEs can be backed out. It also
provides a function to test parameters against critical values, which
is tremendously useful when trying to classify a subject as EUT or RDU.
2017-01-19 21:04:14 +02:00
Brian Albert Monroe 7811428ecb Small update to catch mixture model logic 2017-01-19 19:49:42 +02:00
Brian Albert Monroe 77e28a7049 Update version number, add additional repo for halton 2016-12-20 17:43:53 +02:00
Brian Albert Monroe c9a17f840f Add in option to enforce no correlation for MSL 2016-12-20 16:05:54 +02:00
Brian Albert Monroe b3fbc78780 Fix error with transform from last commit 2016-12-20 11:53:56 +02:00
Brian Albert Monroe bdb67dcd94 Add switch to limit r to < 1 for outcomes == 0 2016-12-20 10:44:04 +02:00
Brian Albert Monroe 1dcd77a851 bump version number 2016-10-29 10:26:10 +02:00
Brian Albert Monroe 583243b0e3 Add options() handling for certain arguments
options() can be set for arguments that already have a default. E.g.
options(MSL.HH = 150) can be set to have all MSL functions use 150
halton draws instead of the default 100.
2016-10-29 10:21:46 +02:00
Brian Albert Monroe 78dad3b3dd Remove unneeded inputParse functions for covar-version 2016-10-29 09:11:40 +02:00
Brian Albert Monroe 4e03de702c Merge branch 'MSL_simplify' 2016-10-28 18:01:44 +02:00
Brian Albert Monroe 16778cc9b5 Get rid of some of the non-MSL.optim functions 2016-10-28 18:01:18 +02:00
Brian Albert Monroe cf4e5dffb5 Update to version 0.1.2 2016-10-28 17:38:13 +02:00
Brian Albert Monroe 293a8a74f6 Update the INST.index to accomodate C++ indexing 2016-10-28 17:37:15 +02:00
Brian Albert Monroe d062dd449e Remove Covar-Version from MSL
Removing Covar-version 1) because I never use it, 2) becuase it allows
me to use the same setup function I use with the RDU functions, which
never had covar-versions. This drops a fair amount of code and creates
very similar MSL functions, so things are now easier to maintain.
2016-10-28 17:25:04 +02:00
Brian Albert Monroe 46641b2e80 Update Description for bugfix 2016-10-28 15:13:56 +02:00
Brian Albert Monroe 2026ce6f36 Update ML and MSL code to correct for RDU issues 2016-10-28 17:14:22 +02:00
Brian Albert Monroe f6e67b9364 Update INVS to match Glenn's 2016-10-26 00:59:43 +02:00
Brian Albert Monroe 64672f092a Update invs to correct for missings 2016-10-25 22:42:27 +02:00
Brian Albert Monroe f5ee78c9fa Update the Documentation for new defaults 2016-10-25 13:21:42 +02:00