bug in arma predict #220

Closed
jseabold opened this Issue Apr 8, 2012 · 1 comment

Comments

Projects
None yet
1 participant
Owner

jseabold commented Apr 8, 2012

Need to look into the problem here.

import numpy as np
import statsmodels.api as sm
from statsmodels.tsa.arima_process import arma_generate_sample
arparams = np.array([.75, -.25])
maparams = np.array([.65, .35])
arparams = np.r_[1, -arparams]
maparam = np.r_[1, maparams]
nobs = 250
y = arma_generate_sample(arparams, maparams, nobs)
import pandas
dates = sm.tsa.datetools.dates_from_range('1980m1', length=nobs)
y = pandas.TimeSeries(y, index=dates)
arma_mod = sm.tsa.ARMA(y, freq='M')
arma_res = arma_mod.fit(order=(2,2), trend='nc', disp=-1)
pred = arma_res.predict(start="1999-8-31", end="2000-8-30")

arma_mod2 = sm.tsa.ARMA(y.values, freq='M')
arma_res2 = arma_mod2.fit(order=(2,2), trend='nc', disp=-1)
pred = arma_res2.predict(start=230, end=248)
Owner

jseabold commented Apr 8, 2012

ARMA results predict just needs a return.

jseabold closed this in 5adfdb2 Apr 8, 2012

@yarikoptic yarikoptic pushed a commit to neurodebian/statsmodels that referenced this issue Apr 25, 2012

@jseabold jseabold BUG: Add return statement to predict. Needs tests! Closes #220. 47fa40f

@yarikoptic yarikoptic added a commit to neurodebian/statsmodels that referenced this issue May 14, 2012

@yarikoptic yarikoptic Merge tag 'v0.4.0rc2' into debian
Version 0.4.0 Release Candidate 2

* tag 'v0.4.0rc2':
  RLS: Set version number to 0.4.0rc2
  ENH: Do not warn on explicit prepend. Closes #217.
  REF/BUG: fix nobs in old example
  REF: sandbox example make rpy optional in garch example
  BUG: fix statsmodels/examples for refactoring bugs, change the run_all.py (glob)
  TST: reduce precision test for breaks_hansen (reported by Yaroslav for Debian)
  BUG: Add return statement to predict. Needs tests! Closes #220.
  ENH add IV2SLS to sandbox sysreg example
  BUG: fix import path in sandbox example
  Docs: add license for qsturng-py and license notice for ordereddict
  DOC: Update stale examples. Closes #219.
  DOC: Fix example
  EX: Remove scikits.ts and larry from example
  EX: Add seed to example and use stationary coefs
  DOC: Fix lag 0 doc. Closes #216.
898303a

@yarikoptic yarikoptic added a commit to neurodebian/statsmodels that referenced this issue Jun 19, 2012

@yarikoptic yarikoptic Merge tag 'v0.4.0' into releases (strategy theirs)
Version 0.4.0

* tag 'v0.4.0': (24 commits)
  DOC: Fix import in CHANGES
  RLS: Set version to 0.4.0
  ENH: expose statsmodels.version.version as statsmodels.__version__
  BUG: Fix import. Closes #229.
  TST: Add test for monotone_fn_inverter
  TST: reduce precision for diagnostic
  ENH: Actually check the args for prepend warning
  EX: Fix refactoring victims
  RLS: Add windows build scripts for bdists.
  RLS: Set version number to 0.4.0rc2
  ENH: Do not warn on explicit prepend. Closes #217.
  REF/BUG: fix nobs in old example
  REF: sandbox example make rpy optional in garch example
  BUG: fix statsmodels/examples for refactoring bugs, change the run_all.py (glob)
  TST: reduce precision test for breaks_hansen (reported by Yaroslav for Debian)
  BUG: Add return statement to predict. Needs tests! Closes #220.
  ENH add IV2SLS to sandbox sysreg example
  BUG: fix import path in sandbox example
  Docs: add license for qsturng-py and license notice for ordereddict
  DOC: Update stale examples. Closes #219.
  ...
e92aa24

@PierreBdR PierreBdR pushed a commit to PierreBdR/statsmodels that referenced this issue Sep 2, 2014

@jseabold jseabold BUG: Add return statement to predict. Needs tests! Closes #220. 5094059
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment