p7]c@s(ddlmZddlmZddlZddlZddlZddlm Z m Z m Z m Z m Z mZmZmZmZmZmZmZddlmZddlmZddljjZddljjZddlm Z dZ!d Z"ej#Z$ej%Z%d ej&fd YZ'd ej(fd YZ)dej*fdYZ+ej,e+e)e-dkr$ddl.j/Z0ddlZe0j1j2j3de4Z5ge5jddgD]4Z6e7e8e6dde7e8e6d^qZ9ej:e5jdddgde9Z;e'e;Z<ndS(i(tlong(tis_numeric_dtypeN( t to_datetimet Int64Indext DatetimeIndextPeriodt PeriodIndext RangeIndext TimestamptSeriestIndext Float64Indext date_ranget period_range(t to_offset(tdata(t ValueWarnings %(model)s Parameters ---------- %(params)s dates : array-like of datetime, optional An array-like object of datetime objects. If a pandas object is given for endog or exog, it is assumed to have a DateIndex. freq : str, optional The frequency of the time-series. A Pandas offset or 'B', 'D', 'W', 'M', 'A', or 'Q'. This is optional if dates are given. %(extra_params)s %(extra_sections)s sTimeseries model base classtTimeSeriesModelcBseZeied6ed6ed6dd6Zd d d ddZd d dZ d dZ d d Z d e d Z d Zd ZeeeZRS(tmodeltparamst extra_paramsttextra_sectionstnonecKs6tt|j||d|||j||dS(Ntmissing(tsuperRt__init__t _init_dates(tselftendogtexogtdatestfreqRtkwargs((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyR.scCsz|dk r|}n |jj}|dkrH|dk rHtdnt}|dk rt|ttfsAytj |}t |st|t st t krt|dt rtdnt|tr|j}nt|}t|tstdn|}WqA|dk r>tdq>qAXnt|ttfr|dkr|jdkr|j}|dk rt}|dk rtjd|tqqn|dk rt|}n|dkr |jdkr |dk rtdqq|dk rv|jdkrvtd|dd |d d |}| rm|j| rmtd n|}q|dk r| r|j|k rtd qq|dk rtdqn|dk }t|ttf}t|t} t|t} |r|jdk nd} tt|jjd} | rT|j| nd} |r|po| po| rtjdtn|r| rtjdtnt}|r| s| r| s| r|}n | }t}||_ ||_!|dk|_"|o | |_#|j#r%|j jnd|_$||_%|j#rI|j nd|j_&|j#rj|j j'nd|j_dS(s Initialize dates Parameters ---------- dates : array_like, optional An array like object containing dates. freq : str, tuple, datetime.timedelta, DateOffset or None, optional A frequency specification for either `dates` or the row labels from the endog / exog data. Notes ----- Creates `self._index` and related attributes. `self._index` is always a Pandas index, and it is always Int64Index, DatetimeIndex, or PeriodIndex. If Pandas objects, endog / exog may have any type of index. If it is an Int64Index with values 0, 1, ..., nobs-1 or if it is (coerceable to) a DatetimeIndex or PeriodIndex *with an associated frequency*, then it is called a "supported" index. Otherwise it is called an "unsupported" index. Supported indexes are standardized (i.e. a list of date strings is converted to a DatetimeIndex) and the result is put in `self._index`. Unsupported indexes are ignored, and a supported Int64Index is generated and put in `self._index`. Warnings are issued in this case to alert the user if the returned index from some operation (e.g. forecasting) is different from the original data's index. However, whenever possible (e.g. purely in-sample prediction), the original index is returned. The benefit of supported indexes is that they allow *forecasting*, i.e. it is possible to extend them in a reasonable way. Thus every model must have an underlying supported index, even if it is just a generated Int64Index. s,Frequency provided without associated index.isNumeric index givensCould not coerce to date indexs2Non-date index index provided to `dates` argument.sMNo frequency information was provided, so inferred frequency %s will be used.sYNo frequency information was provided with date index and no frequency could be inferred.tstarttendiR sEThe given frequency argument could not be matched to the given index.sBThe given frequency argument is incompatible with the given index.sKGiven index could not be coerced to dates but `freq` argument was provided.sLAn unsupported index was provided and will be ignored when e.g. forecasting.s|A date index has been provided, but it has no associated frequency information and so will be ignored when e.g. forecasting.N((tNoneRt row_labelst ValueErrortFalset isinstanceRRtnptasarrayRR ttupletfloatR tvaluesRR R t inferred_freqtTruetwarningstwarnRRR tequalsRRtrangeRtshapet_indext_index_generatedt _index_nonet _index_datest _index_freqt_index_inferred_freqRtfreqstr(RRR tindexR.R5tresampled_indext has_indext date_indext int_indext range_indexthas_freqt incrementt is_incrementtindex_generated((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyR6s+                            cCs|d kr|j}n|}t|ttf}t|t}t|t}t|}t|}|r/t|t t t j fr/|dkr| |kr||}q/||dkr/y|j } |j} Wn#tk r|j} |j} nX| |d| } td| d| d| }q/n| r|r| rt|t t t j fr|dkr| |kr||}q||dkrtt j|dt |d}qn|r)|tkrt} nt} t|t t t j fr|dkr&| |kr&|||}q&|t|dkru| d|ddt |dd|j}|d}q&||}q)|tkrt|d|j} n t|} | |dkr)| d|dd | d|j}|d| ks&| d|ddt|dd|j}q&q)n|rA|j|}n^|sM|ry ||Wn+ttfk r}tt|nX|}n|j|}||k }t|tr|j}n|}|||d |fS( s4 Get the location of a specific key in an index Parameters ---------- key : label The key for which to find the location if the underlying index is a DateIndex or a location if the underlying index is a RangeIndex or an Int64Index. base_index : pd.Index, optional Optionally the base index to search. If None, the model's index is searched. Returns ------- loc : int The location of the key index : pd.Index The index including the key; this is a copy of the original index unless the index had to be expanded to accomodate `key`. index_was_expanded : bool Whether or not the index was expanded to accomodate `key`. Notes ----- If `key` is past the end of of the given index, and the index is either an Int64Index or a date index, this function extends the index up to and including key, and then returns the location in the new index. iiR"tstoptstepitperiodsR R#N(R$R5R(RRRRttypetlentintRR)tintegerR"RGtAttributeErrort_startt_steptarangeR R R RRtget_loct IndexErrorR&tKeyErrortstrtsliceRF(Rtkeyt base_indexR<R?R@RAt index_classtnobstbase_index_starttbase_index_stepRFtindex_fntdate_keytloctetindex_was_expandedR#((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyt_get_index_locs    !        ,             cCsy|j||\}}}Wntk r }yt|tttjfsg|jjj |}nt|t r|j }nt|tj r|j tkrtj|}q|d}nt|tjsn|jj|d }t}Wq |q XnX|||fS(sy Get the location of a specific key in an index or model row labels Parameters ---------- key : label The key for which to find the location if the underlying index is a DateIndex or is only being used as row labels, or a location if the underlying index is a RangeIndex or an Int64Index. base_index : pd.Index, optional Optionally the base index to search. If None, the model's index is searched. Returns ------- loc : int The location of the key index : pd.Index The index including the key; this is a copy of the original index unless the index had to be expanded to accomodate `key`. index_was_expanded : bool Whether or not the index was expanded to accomodate `key`. Notes ----- This method expands on `_get_index_loc` by first trying the given base index (or the model's index if the base index was not given) and then falling back to try again with the model row labels as the base index. ii(RaRSR(RKRR)RLRR%RQRUR"tndarraytdtypetbooltargmaxtnumberstIntegralR'(RRVRWR^R<R`R_((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyt_get_index_label_locs(    c Csy|j|\}}}Wntk r;tdnX|d krgt|t|jd}ny|j|\}}}Wntk rtdnXt|tr|j}nt|tr|j d}n||} ||krt dn|d k rtt| t|ks5t dnt|j t j  re| ret jdtnt|} nz|jr|j r|j jd k r|p| r|j j||d!} q|st jdtqn|jrd } n| d k r)| d|j _| d |j _| |j _n$d |j _d |j _d |j _t|j} t|| dd} || 8}||| | fS( s Get the location of a specific key in an index or model row labels Parameters ---------- start : label The key at which to start prediction. Depending on the underlying model's index, may be an integer, a date (string, datetime object, pd.Timestamp, or pd.Period object), or some other object in the model's row labels. end : label The key at which to end prediction (note that this key will be *included* in prediction). Depending on the underlying model's index, may be an integer, a date (string, datetime object, pd.Timestamp, or pd.Period object), or some other object in the model's row labels. index : pd.Index, optional Optionally an index to associate the predicted results to. If None, an attempt is made to create an index for the predicted results from the model's index or model's row labels. silent : bool, optional Argument to silence warnings. Returns ------- start : integer The index / observation location at which to begin prediction. end : int The index / observation location at which to end in-sample prediction. The maximum value for this is nobs-1. out_of_sample : int The number of observations to forecast after the end of the sample. prediction_index : pd.Index or None The index associated with the prediction results. This index covers the range [start, end + out_of_sample]. If the model has no given index and no given row labels (i.e. endog/exog is not Pandas), then this will be None. Notes ----- The arguments `start` and `end` behave differently, depending on if they are integer or not. If either is an integer, then it is assumed to refer to a *location* in the index, not to an index value. On the other hand, if it is a date string or some other type of object, then it is assumed to refer to an index *value*. In all cases, the returned `start` and `end` values refer to index *locations* (so in the former case, the given location is validated and returned whereas in the latter case a location is found that corresponds to the given index value). This difference in behavior is necessary to support `RangeIndex`. This is because integers for a RangeIndex could refer either to index values or to index locations in an ambiguous way (while for `Int64Index`, since we have required them to be full indexes, there is no ambiguity). sYThe `start` argument could not be matched to a location related to the index of the data.isWThe `end` argument could not be matched to a location related to the index of the data.s)Prediction must have `end` after `start`.seInvalid `index` provided in prediction. Must have length consistent with `start` and `end` arguments.sBecause the model data (`endog`, `exog`) were not given as Pandas objects, the prediction output will be Numpy arrays, and the given `index` argument will only be used internally.smNo supported index is available. Prediction results will be given with an integer index beginning at `start`.iiN(RhRSR$tmaxRJR5R(RUR"RFR&Rt PandasDataR0R1RR R6R7R%t predict_startt predict_endt predict_datesR( RR"R#R<tsilentt start_indext start_oost end_indextend_oostprediction_indexRYt out_of_sample((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyt_get_prediction_indexsV<                  cCs |jjS(N(Rtxnames(R((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyt_get_exog_namesUscCs+t|ts|g}n||j_dS(N(R(tlistRRv(Rtvals((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyt_set_exog_namesXs N(t__name__t __module__t_tsa_doct _model_doct_generic_paramst_missing_param_doct__doc__R$RRRaRhR'RuRwRztpropertyt exog_names(((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyR(s    I  tTimeSeriesModelResultscBseZddZRS(g?cCs/|j|_tt|j||||dS(N(RRRR(RRRtnormalized_cov_paramstscale((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyRbs (R{R|R(((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyRastTimeSeriesResultsWrappercBsKeZiZejejjeZidd6Zejejj eZ RS(Rtpredict( R{R|t_attrstwrapt union_dictstbasetLikelihoodResultsWrappert _wrap_attrst_methodst _wrap_methods(((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyRhs   t__main__t as_pandastyeartquarterit:itrealgdptrealinvtrealconsR<(=tstatsmodels.compat.pythonRtstatsmodels.compat.pandasRRfR0tnumpyR)tpandasRRRRRRRR R R R R tpandas.tseries.frequenciesRtstatsmodels.baseRtstatsmodels.base.modelRRtstatsmodels.base.wrappertwrapperRtstatsmodels.tools.sm_exceptionsRR}R~t_model_params_docRRtLikelihoodModelRtLikelihoodModelResultsRtResultsWrapperRtpopulate_wrapperR{tstatsmodels.apitapitsmtdatasetst macrodatatloadR'tmdatatxRTRKRt DataFrametdftex_mod(((s=lib/python2.7/site-packages/statsmodels/tsa/base/tsa_model.pyts:   R  ;   K%