Gmm new rebased #1258

Merged
merged 36 commits into from Dec 20, 2013

Projects

None yet

2 participants

@josef-pkt
Member

just a rebased version of #1105 BUG/ENH IV2SLS GMM

josef-pkt added some commits Oct 4, 2013
@josef-pkt josef-pkt BUG: fix predict (was refactoring victim) dc76f7e
@josef-pkt josef-pkt ENH: GMM fitgmm allow choice of optimizer 81aa16d
@josef-pkt josef-pkt REF: gmm, partway through refactoring cbec127
@josef-pkt josef-pkt ENH: first try at CU-GMM 4b04783
@josef-pkt josef-pkt REF: use pinv for weight matrix in fititer 34ba83d
@josef-pkt josef-pkt BUG/REF: gmm cleanup, with unittests, fititer compares to Stata f3f8cc6
@josef-pkt josef-pkt REF: gmm cov use ddof=0 as default, more options, OLS test case f4311a7
@josef-pkt josef-pkt BUG/TST: IV2SLS fix df_resid ("small"), plus unit tests b7d5ef9
@josef-pkt josef-pkt TST: IV2SLS add df_model, more unit test 47cb2fb
@josef-pkt josef-pkt ENH/BUG: add IVRegressionResults, fix fvalue, see #1104 67dc589
@josef-pkt josef-pkt REF/BUG: move spec_hausman to results class, fix it 1baaf1d
@josef-pkt josef-pkt ENH: IVRegressionResults add summary b87b639
@josef-pkt josef-pkt REF/ENH GMM, IVGMM: add summary, partial cleanup weights/inv_weights 3bdfad1
@josef-pkt josef-pkt REF/TST trying to replicate ivreg2, bse in final iteration? b567b90
@josef-pkt josef-pkt TST: compare with Stata oneway, twoway, oneway still does something d…
…ifferent
830ae97
@josef-pkt josef-pkt ENH: add HAC for weightmatrix, fix arguments, add compare_j (untested) dad6cd6
@josef-pkt josef-pkt BLD: add GMM test dta to setup.py (file should be moved) 37b55c1
@josef-pkt josef-pkt BLD: try again, add test data file to manifest.in 6e0e5f4
@josef-pkt josef-pkt TST: add (left out) results file 4c525e1
@josef-pkt josef-pkt REF/BUG: add `disp` option, fix missing nobs fa287c5
@josef-pkt josef-pkt ENH: multivariate_tools in stats (verified with Stata but no unittest…
…s yet)
35c5519
@josef-pkt josef-pkt BUG/REF: fix wrong keyword (cov/momcov), add wargs "centered"
         fix tests to use centered keyword (Stata has non-centered default)
6bdc342
@josef-pkt josef-pkt REF: streamline argument/option handling, rename keywords 6c56fe7
@josef-pkt josef-pkt DOC/BUG: misc fixes for unused, add comments and doc for options 6b7ef4d
@josef-pkt josef-pkt REF/ENH: add start_weights method, untested: add LinearIVGMM, other ENH 93854e1
@josef-pkt josef-pkt TST: add test with options corresponding to Stata defaults (higher ag…
…reement)
6d3090e
@josef-pkt josef-pkt BUG/TST: fix enh LinearIVGMM and NonlinearIVGMM, unittests 1afee3d
@josef-pkt josef-pkt REF/ENH: make weight_method='iid' work, gmm replicates ols b4f314a
@josef-pkt josef-pkt ENH: finish CUE, tried out on example, no unittests yet f875bed
@josef-pkt josef-pkt BUG/REF: fix sign of score, fix start_weights 1b820d2
@josef-pkt josef-pkt TST add Poisson/Exponential test case 8e38829
@josef-pkt josef-pkt TST add unittest for j_test, reenable summary() smoketest 0b5a03d
@josef-pkt josef-pkt ENH: inherit from Model for data handling, temporary fix for xnames f8f0bd3
@josef-pkt josef-pkt STY move example code from module 1d7a0a8
@josef-pkt josef-pkt REF: fix to get old examples to run, DistQuantilesGMM needs refactoring 69286f1
@josef-pkt josef-pkt BUG: pylint found errors 0a7a9b9
@coveralls

Coverage Status

Coverage remained the same when pulling 0a7a9b9 on josef-pkt:gmm_new_rebased into 5a731a3 on statsmodels:master.

@josef-pkt
Member

TravisCI is green, I will merge this soon.

@josef-pkt josef-pkt merged commit 8b0f906 into statsmodels:master Dec 20, 2013

1 check passed

default The Travis CI build passed
Details
@josef-pkt josef-pkt deleted the josef-pkt:gmm_new_rebased branch Dec 20, 2013
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment