Issue Labels

[jbrockmendel]

Labels to consider implementing:

• CI
• EX (Example)
• SUMM
• Something for "consider implementing FooBar model"
• Segfaults (severe, un-recoverable bugs)
• Internal consistency (params is a method of emplikeAFT in aft_el.py, not an attribute #3629, orig_endog vs endog_lagged vs... #3689, SUMM: Topic Week - Internal Consistency #4030, REF: consistent naming: results_frame, summary_frame, table #4522, REF/Design: should families be stateless #4742...)
• Overflows/Numeric
• Failing Test (Very many tests (86 errors, 5 failures) do not pass  #4850)/Inconsistency on Windows (windows and linux different #4737)/...
• Comment/Question (Status of resampling methods in statsmodels #4956, The APIs confused me when I wanted to forecast with ARMA #4961)
• Regression (might want a more specific name for this context; e.g. Issue with lilliefors after upgrading to 0.9.0 from 0.8.0 (regression?) #5333)

A motivated maintainer can accomplish immediately:

• Cleanup: Remove branches from main repo #1280: Remove branches from main repo

@bashtage collecting these here instead of pinging you a zillion times. I've looked through about half the Issues so far, will post below if I identify more

• Roadmap release 0.11 #5406: RLS
• Statsmodels 0.10.0 #5330: RLS
• conda can't get statsmodels=0.8.0_dev #3982: RLS -- not sure about this label
• REL: release 0.8 - changes 0.8rc1 #3075: RLS
• release 0.9 #3076: RLS
• roadmap 0.9 #3431: RLS
• Roadmap release 0.10 #4507: RLS
• MAINT: release 0.9rc1 to 0.9 final #4573: RLS
• MAINT: Release 0.10 #5290: RLS
• Import in Python 2.7.15 fails with undefined symbol: PyUnicodeUCS4_DecodeUTF8 #5371: BLD
• ValueError: Unknown MS Compiler version 1914 #5035: BLD
• cl.exe failing when installing statsmodel #4407: BLD
• error: Microsoft Visual C++ 14.0 is required #4160: BLD
• RuntimeError: Running cythonize failed! #4150: BLD
• ** On entry to DLASCL parameter number 4 had an illegal value Crashes Python #3596: BLD
• Please declare dependencies at least for the wheels #3373: BLD
• Tracking issue for pip install test failures #4657: BLD
• DESIGN: Speed up innovations algorithm / filter with Cython or Numba? #5326: PERF
• Speed up the cross-correlation function #5199: PERF
• ENH: speeding up model training by moving it to GPU? #4752: PERF
• BUG: import statsmodels slow and loads a lot, including pandas #3082: PERF (closeable?)
• Design/FAQ-D: global memoize #3237: PERF
• ENH/Performance: Binning for local nonparametric estimation #3399: PERF
• adfuller unexpectedly using all available cores #3542: PERF
• Lazy loading of data for model #3609: PERF
• KernelReg performance in a for loop #4187: PERF
• ENH/Design using numba #4239: PERF --> already has tag, but I think duplicate of something else above
• REF: VAR IRF avoid repeated run of MC error bands #4546: PERF
• ENH: GLM.fit_regularized needs better default_params #4608: PERF
• controlling number of cores #5344: PERF
• timing of wrapper code (pandas integration) #89: PERF (likely closable)
• Performance: sequential regression #296: PERF
• REF: use pandas categoricals instead of np.unique #2360: PERF
• memory for FDR #2447: PERF
• ENH: out of core GAM #2585: PERF
• FAQ: number of cores used in statsmodels, and how to restrict it #2914: PERF
• FAQ: Performance tips #1663: PERF
• REF: emplike OriginResults repeated calculation #1848: PERF
• smaller memory footprint of models #2113: PERF
• DEPR: Is sandbox.formula supplanted by patsy? #5688: BUG
• RegressionResults.__str__ returns None #4618: BUG
• MNLogit.from_formula loses labels in categorical #4544: BUG, pandas-compat
• MultinomialResults.predict() doesn't wrap pandas #4541: BUG, pandas-compat
• Incorrect error message in tools.tool? #4332: BUG (not sure about label, CLN doesnt seem right)
• is sm.open_help ever defined? #5134: CLN
• Identical strip_rc defined in setup, tools.testing, used nowhere #5126: CLN
• gof.gof_binning_discrete "# copy/paste, remove code duplication when it works" #5125: CLN
• README_l1.txt references non-existent files #5111: CLN
• imports in both iolib.__init__ and iolib.api #5062: CLN
• “Copied from PR #2380 remove when merged” #5023: CLN
• _diagnostic_count functions starting with "test_" should renamed #5019: CLN
• De-duplicate HoldIt class #4994: CLN
• LikelihoodModel "copied from model.py" should be removed from garch #4422: CLN
• #delete when testing is finished #4419: CLN
• omni_normtest TODO ntoe #4062: CLN
• KeyError? #4110: CLN
• vector_ar ... pandas.Panel is deprecated #5025: CLN
• REF: Kernel density estimation cleanup and improvement #4220: CLN, REF, ENH
• R code broken in Linear Mixed Effects example notebook #5225: EX
• examples files un-runnable #5037: EX
• Bug? initial conditions in statespace_local_linear_trend.ipynb #2326: EX
• skip_one set correctly in IRLS tests? #4935: TST
• Hypothesis for testing #3078: TST, ENH
• TST: Codecov False-Positives Caused by Noise in mixed_linear_model tests? #5082: TST
• Audit Test Suite #4966: TST
• summary() unit tests mock date time #1376: TST
• REF: Wrapping "columns"/"columns_eq"/"rows"/"rows_eq" #4549: pandas-wrapping
• MultinomialResults.predict() doesn't wrap pandas #4541: pandas-wrapping
• Marginal Effects need to be wrapped #375: pandas-wrapping
• MNLogit.cov_params is broken when wrapped. #814: pandas-wrapping
• BUG: VAR cov_params pandas wrapper broken #2270: pandas-wrapping
• categorical does not work with Pandas DataFrames #1342: pandas-categorical
• MNLogit.from_formula loses labels in categorical #4544: pandas-categorical
• BLD: Replace LGTM with Codacy #4894: pandas-compat
• ENH: out of core GAM #2585: pandas-compat
• ANOVA documentation clarity. #3306: DOC
• DOC: Confusion in power functions and documentation #4085: DOC
• Incorrect sentence in documentation on endogeneity #4199: DOC
• REF: Removal of deprecated objects for 0.9 #3433: DEPR
• REF deprecate lower case link classes #3462: DEPR
• DEPR: Deprecations from prior versions #5092: DEPR
• Deprecate multiple-return signatures #4964: DEPR/DES
• _diagnostic_count functions starting with "test_" should renamed #5019: MAINT
• MAINT: python 2.7 specific subclasses of ResultsWrappers #5472: py27-specific
• numerical derivative precision #583: NumericalPrecision
• BUG: Kalman filter with failures with single precision input #2262: NumericalPrecision
• REF/Style: avoid Stata names #2484: Naming Conventions/Internal Consistency
• REF: change wendog -> transformed_endog or something perhaps model-specific #2314: Naming Conventions/Internal Consistency
• consistent definitions of nobs, n_xxx, k_xxx #1664: Naming Conventions/Internal Consistency
• Trend in tsa models: standardize #2207: Naming Conventions/Internal Consistency
• log likelihood vs. negative log likelihood, score vs. negative score #1162: Naming Conventions/Internal Consistency
• Use abstract base classes to enforce model conventions #1181: Naming Conventions/Internal Consistency
• names in robust, rename, standardize #1335: Naming Conventions/Internal Consistency
• Style: Naming Conventions #487: Naming Conventions/Internal Consistency
• rename rlm.M #73: Naming Conventions/Internal Consistency
• standardizing names #157: Naming Conventions/Internal Consistency

Closeable? (see also #5155):

• can't run the latest master build, also would like to submit changes, how? #3555
• old module info.py #3214
• FAQ-Maintainer windows build scripts removal #3780
• anaconda3-4.2.0+python3.5 +ImportError: cannot import name 'getargspec' #4466
• DOC: mixed effects models example fails to import rpy2 #4204 Closable
• genfromdta does not accept file:// URI schemes. #475: genfromdta
• genfromdta and unicode #476: genfromdta
• subset semantics/syntax #469
• code convention and pylint #695: pylint
• what's DynamicVAR #780
• tools for code maintenance #829
• formula.api cleanup remove classes #833: closed by Clean up the smf namespace  #5146
• need central version check for dependency #903: closed by Jbrockmendel pversion #5592
• improve cache decorator #886
• Remove version compatibility code #1063
• old branch misc_io #1237
• FAQ: using sympy for statistics #2463
• ENH: tsa helper function: innovation algorithm for MA #2593: closed by ENH: Add innovations algorithm #5042
• TST: occasional failure in test_numdiff.TestGradLogit.test_hess #1893
• ENH: zero-inflated poisson #1086
• string versus bytes, py 3 conversion/compatibility #1324: str/bytes
• ENH: Add seasonal plotting. #1290: unicode
• TST: missing unicode tests ? #1746: unicode
• TST: safeguard against multiprocessing problems, limit number of cores #1892
• ENH: complex step derivative #2341
• rpy2 stop-gaps? #1557 discussion of replacing rpy2, should be long-defunct
• Add GARCH + plus flexible enhancements #437 - close as duplicate of availability of ARCH and GARCH #5441, which has the FAQ label
• Basic ARCH(q) model for autoregressive Time series #1270 - close as duplicate of availability of ARCH and GARCH #5441, which has the FAQ label

[bashtage]

I closed what I felt comfortable with.