p7]c@s4dZddlmZddlZddlZddlZddlZddlmZddl Z ddl m Z ddl mZddlmZmZmZmZdd gZd.add/dYZeZidd6dd6Zide6de6dd6ZdZdedZddZdZdeedZ dZ!dZ"dZ#dZ$d Z%d!Z&d"Z'd#e(fd$YZ)d%e)fd&YZ*d'Z+d0d1dddeedeeddeded*Z,d2d3dddeedddeded+ Z-d,e(fd-YZ.dS(4s Run x12/x13-arima specs in a subprocess from Python and curry results back into python. Notes ----- Many of the functions are called x12. However, they are also intended to work for x13. If this is not the case, it's a bug. i(tprint_functionN(twarn(t iteritems(tBunch(tX13NotFoundErrort IOWarningtX13Errort X13Warningtx13_arima_select_ordertx13_arima_analysiss x13as.exetx13assx12a.exetx12at_freq_to_periodcBseZdZRS(cCs=|jdrdS|jdr&dS|jdr9dSdS(NtMi tQitWi4(t startswith(tselftkey((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt __getitem__s (t__name__t __module__R(((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyR sR i RitlogtnonetautocCs|r dSdS(Ntyestno((tx((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt,tcCs-|dk r0|jtr0tjj|}n|stddda|dkrjtjdd}n|stjdd}qn<|dkrtjdd}|stjdd}qnxbtD]V}tjj||}y'tj |dtj dtj |SWqt k r qXqWt SdS(s If x12path is not given, then either x13as[.exe] or x12a[.exe] must be found on the PATH. Otherwise, the environmental variable X12PATH or X13PATH must be defined. If prefer_x13 is True, only X13PATH is searched for. If it is false, only X12PATH is searched for. NitX12PATHRtX13PATHtstdouttstderr( tNonetendswitht _binary_namestostpathtdirnametgetenvtjoint subprocesst check_calltPIPEtOSErrortFalse(tx12patht prefer_x13tbinarytx12((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt _find_x12/s*     cCs%t|}|s!tdn|S(Nsxx12a and x13as not found on path. Give the path, put them on PATH, or set the X12PATH or X13PATH environmental variable.(R3R(R/((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt _check_x12Ss cCsetjd|}d}t|dkrEt||\}}n||d}d}||fS(s Takes something like (1 1 0)(0 1 1) and returns a arma order, sarma order tuple. Also accepts (1 1 0) and return arma order and (0, 0, 0) s \([0-9 ]*?\)cSs+ttttjdd|jdS(Ns[()]Rt (ttupletmaptinttretsubtsplit(R((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pytcleancsii(iii(R9tfindalltlenR7(torderR<tsorder((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt _clean_order\s cCs|r|rtdn|r4|d|g}n%|rM|d|g}n ||g}|ro||g7}ntj|dtjdtjS(Ns&Cannot specify both meta and datameta.s-m s-d R R!(t ValueErrorR*tPopenR,tSTDOUT(R/tspecpathtoutnametmetatdatametatargs((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pytrun_specos  cCssd}|dj|d|d7}|dk rQ|dj|d|d7}n|dj|d|d7}|S(Ns smaxorder = ({0} {1}) iismaxdiff = ({0} {1}) sdiff = ({0} {1}) (tformatR"(tmaxordertmaxdifftdifftoptions((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt_make_automdl_optionss  !cCst|dr|j}n't|dr6|j}n tdydj|}WnNtk rddlm}|jdkrd}qdj||}nX|S( Ntnametcolumnss0exog is not a Series or DataFrame or is unnamed.R5i(t_make_exog_namesitx1( thasattrRQRRRBR)t TypeErrortstatsmodels.base.dataRStndim(texogt var_namesRS((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt_make_var_namess     cCs| r|dkrdSd}|r0|d7}n|dk rt|}|dj|7}|djdjtt|jjj7}n|d7}|S(NRs regression{ s variables = (td) s user = ({0}) s data = ({0}) s s} ( R"R[RKR)R7tstrtvaluestravelttolist(ttradingRYtreg_specRZ((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt_make_regression_optionss   " cCs-|dkrdSd}|dj|7}|S(NRs forecast{ smaxlead = ({0}) }} (R"RK(tforecast_yearst forecast_spec((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt_make_forecast_optionss  cCsd||jddj}|r>d|kr>t|n"|r`d|kr`t|tndS(Nsspc:itERRORtWARNING(tfindtstripRRR(terrors((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt _check_errorss c Csnddlm}ddlm}|||dddd ddd d }|j|jd i|d 6|S(si Convert x to a DataFrame where x is a string in the format given by x-13arima-seats output. i(tStringIO(tread_csvtskiprowsitheadertseps tenginetpythonRRiN(tstatsmodels.compatRltpandasRmR"t set_indextrename(RtdatesRQRlRmtout((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt_convert_out_to_seriess cCs(t|d}|j}WdQX|S(Ntr(topentread(tfnametfintfout((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt_open_and_readstSpeccBs)eZedZdZdZRS(cCs|jjjddS(NRR(t __class__Rtreplace(R((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt spec_namescKs"d}|jd|jd|jS(Ns/{name} {{ {options} }} RQRO(RKRRO(Rtkwargstspec((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt create_specscKs]d}xGt|D]9\}}|dj||7}|jji||6qW||_dS(NRs{0}={1} (RRKt__dict__tupdateRO(RRRORtvalue((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt set_optionss (RRtpropertyRRR(((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyRs t SeriesSpeccBsMeZdZdeed ddd ddggd d dd d dddZRS( sh Parameters ---------- data appendbcst : bool appendfcst : bool comptype compwt decimals modelspan name period precision to_print to_save span start title type Notes ----- Rarely used arguments divpower missingcode missingval saveprecision trimzero sUnnamed Seriesiii Riayiʚ;cCsytt||g\}}dj|d }dj|d }|jd|d|d|d| d|d |d |dS( Ns"{0}"i@iOtdatat appendbcstt appendfcsttperiodtstartttitleRQ(R7t_bool_to_yes_noRKR(RRRQRRtcomptypetcompwttdecimalst modelspanRt precisiontto_printtto_savetspanRRt series_typetdivpowert missingcodet missingvalt series_name((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt__init__s N(((ii(RRt__doc__R.R"R(((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyRs  c Cs}t|drGt|jdkr3tdn||jd}ndjdjtt|jj }yt |j j }Wn:t tfk rddlm}t ||j }nX|j d}|d kr|j|j}}n.|d kr|j|j}}n td t|d r@|jp:d }nd }td|d |d|d|ddj||}|S(NRRis3Does not handle DataFrame with more than one columnis({0})s i(t infer_freqi isqOnly monthly and quarterly periods are supported. Please report or send a pull request if you want this extended.RQsUnnamed SeriesRRRRs{0}.{1}(RUR>RRRBRKR)R7R\R]R_R tindextfreqstrtAttributeErrortpandas.tseries.apiRtyeartmonthtquarterRQR( RRRRt start_dateRtstperiodRQt series_spec((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pytpandas_to_series_specs.*     iicCs~t|}t|tjtjfs| dks?| dkrNtdntj|dtjd| dt|d| }nt |}|j }|dj t |7}|r|d7}nt |||}|dj |7}|t||7}|t|7}|d 7}| r"|Stjd td d }tjd t}z|j|jd |j|jt||jd |j}|j|jj}| rt|jjnt|jd}t|t|jd}t|jd}t|jd}t|jd}Wdy$tj|jtj|jWnst k rtj!j"|jrt#dj |jt$ntj!j"|jrt#dj |jt$qnXXt%||j&d}t%||j&d}t%||j&d}| sGt'd|d|d|d|d|d|}n3t'd|d|d|d|d|d|d|}|S(s Perform x13-arima analysis for monthly or quarterly data. Parameters ---------- endog : array-like, pandas.Series The series to model. It is best to use a pandas object with a DatetimeIndex or PeriodIndex. However, you can pass an array-like object. If your object does not have a dates index then ``start`` and ``freq`` are not optional. maxorder : tuple The maximum order of the regular and seasonal ARMA polynomials to examine during the model identification. The order for the regular polynomial must be greater than zero and no larger than 4. The order for the seaonal polynomial may be 1 or 2. maxdiff : tuple The maximum orders for regular and seasonal differencing in the automatic differencing procedure. Acceptable inputs for regular differencing are 1 and 2. The maximum order for seasonal differencing is 1. If ``diff`` is specified then ``maxdiff`` should be None. Otherwise, ``diff`` will be ignored. See also ``diff``. diff : tuple Fixes the orders of differencing for the regular and seasonal differencing. Regular differencing may be 0, 1, or 2. Seasonal differencing may be 0 or 1. ``maxdiff`` must be None, otherwise ``diff`` is ignored. exog : array-like Exogenous variables. log : bool or None If None, it is automatically determined whether to log the series or not. If False, logs are not taken. If True, logs are taken. outlier : bool Whether or not outliers are tested for and corrected, if detected. trading : bool Whether or not trading day effects are tested for. forecast_years : int Number of forecasts produced. The default is one year. retspec : bool Whether to return the created specification file. Can be useful for debugging. speconly : bool Whether to create the specification file and then return it without performing the analysis. Can be useful for debugging. start : str, datetime Must be given if ``endog`` does not have date information in its index. Anything accepted by pandas.DatetimeIndex for the start value. freq : str Must be givein if ``endog`` does not have date information in its index. Anything accapted by pandas.DatetimeIndex for the freq value. print_stdout : bool The stdout from X12/X13 is suppressed. To print it out, set this to True. Default is False. x12path : str or None The path to x12 or x13 binary. If None, the program will attempt to find x13as or x12a on the PATH or by looking at X13PATH or X12PATH depending on the value of prefer_x13. prefer_x13 : bool If True, will look for x13as first and will fallback to the X13PATH environmental variable. If False, will look for x12a first and will fallback to the X12PATH environmental variable. If x12path points to the path for the X12/X13 binary, it does nothing. Returns ------- res : Bunch A bunch object with the following attributes: - results : str The full output from the X12/X13 run. - seasadj : pandas.Series The final seasonally adjusted ``endog`` - trend : pandas.Series The trend-cycle component of ``endog`` - irregular : pandas.Series The final irregular component of ``endog`` - stdout : str The captured stdout produced by x12/x13. - spec : str, optional Returned if ``retspec`` is True. The only thing returned if ``speconly`` is True. Notes ----- This works by creating a specification file, writing it to a temporary directory, invoking X12/X13 in a subprocess, and reading the output directory, invoking exog12/X13 in a subprocess, and reading the output back in. s=start and freq cannot be none if endog is not a pandas objectRRtperiodstfreqstransform{{function={0}}} s outlier{} sautomdl{{{0}}} sx11{ save=(d11 d12 d13) }tdeletetsuffixs.spctutf8is.errs.outs.d11s.d12s.d13NsFailed to delete resource {0}tseasadjttrendt irregulartobservedtresultsR R((R4t isinstancetpdt DataFrametSeriesR"RBt DatetimeIndexR>RRRKt _log_to_x12RPRbRettempfiletNamedTemporaryFileR.twritetencodetcloseRJRQtwaitR R|tprintRRkR%tremoveR-R&texistsRRRyRtX13ArimaAnalysisResult(tendogRLRMRNRYRtoutlierR`RctretspectspeconlyRRt print_stdoutR/R0tspec_objRROtftempintftempouttpR RjRRRRtres((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyR @st^                cCst|d| d|d|d|d|d|d|d|d |d | d | d |  }tjd |j}|j}tjd|jrt}n$tjd|jrt}nt}t|\}}td|d|d|d|jd|j }|S(sy Perform automatic seaonal ARIMA order identification using x12/x13 ARIMA. Parameters ---------- endog : array-like, pandas.Series The series to model. It is best to use a pandas object with a DatetimeIndex or PeriodIndex. However, you can pass an array-like object. If your object does not have a dates index then ``start`` and ``freq`` are not optional. maxorder : tuple The maximum order of the regular and seasonal ARMA polynomials to examine during the model identification. The order for the regular polynomial must be greater than zero and no larger than 4. The order for the seaonal polynomial may be 1 or 2. maxdiff : tuple The maximum orders for regular and seasonal differencing in the automatic differencing procedure. Acceptable inputs for regular differencing are 1 and 2. The maximum order for seasonal differencing is 1. If ``diff`` is specified then ``maxdiff`` should be None. Otherwise, ``diff`` will be ignored. See also ``diff``. diff : tuple Fixes the orders of differencing for the regular and seasonal differencing. Regular differencing may be 0, 1, or 2. Seasonal differencing may be 0 or 1. ``maxdiff`` must be None, otherwise ``diff`` is ignored. exog : array-like Exogenous variables. log : bool or None If None, it is automatically determined whether to log the series or not. If False, logs are not taken. If True, logs are taken. outlier : bool Whether or not outliers are tested for and corrected, if detected. trading : bool Whether or not trading day effects are tested for. forecast_years : int Number of forecasts produced. The default is one year. start : str, datetime Must be given if ``endog`` does not have date information in its index. Anything accepted by pandas.DatetimeIndex for the start value. freq : str Must be givein if ``endog`` does not have date information in its index. Anything accapted by pandas.DatetimeIndex for the freq value. print_stdout : bool The stdout from X12/X13 is suppressed. To print it out, set this to True. Default is False. x12path : str or None The path to x12 or x13 binary. If None, the program will attempt to find x13as or x12a on the PATH or by looking at X13PATH or X12PATH depending on the value of prefer_x13. prefer_x13 : bool If True, will look for x13as first and will fallback to the X13PATH environmental variable. If False, will look for x12a first and will fallback to the X12PATH environmental variable. If x12path points to the path for the X12/X13 binary, it does nothing. Returns ------- results : Bunch A bunch object that has the following attributes: - order : tuple The regular order - sorder : tuple The seasonal order - include_mean : bool Whether to include a mean or not - results : str The full results from the X12/X13 analysis - stdout : str The captured stdout from the X12/X13 analysis Notes ----- This works by creating a specification file, writing it to a temporary directory, invoking X12/X13 in a subprocess, and reading the output back in. R/RYRRR`RcRLRMRNRRR0s&(?<=Final automatic model choice : ).*sMean is not significanttConstantR?R@t include_meanRR ( R R9tsearchRtgroupR.tTrueRARR (RRLRMRNRYRRR`RcRRRR/R0RtmodelR?RR@R((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyRs"S      RcBseZdZdZRS(cKs1x*t|D]\}}t|||q WdS(N(Rtsetattr(RRRR((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyRQscCsddlm}|}|jdddt\}}|jjd|ddt|djd |jjd|ddt|djd |j jd|d dt|d jd |j jd|d dt|d jd|j |S(Ni(t _import_mpliitsharextaxitlegendtObservedsSeas. AdjusteditTrendit Irregular( tstatsmodels.graphics.utilsRtsubplotsRRtplotR.t set_ylabelRRRt tight_layout(RRtplttfigtaxes((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyRUs  (RRRR(((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyRPs (s x13as.exeR sx12a.exeR ((ii(ii(ii(ii(/Rt __future__RR%R*RR9twarningsRRtRtstatsmodels.compat.pythonRtstatsmodels.tools.toolsRtstatsmodels.tools.sm_exceptionsRRRRt__all__R$R t_period_to_freqRR.R"RRR3R4RARJRPR[RbReRkRyRtobjectRRRR RR(((s2lib/python2.7/site-packages/statsmodels/tsa/x13.pyt sT     "   $      3 '      c