ó áp7]c!@s”dZddlmZmZmZddlZddlmZ ddl m Z m Z ddl mZddlmZd „Zd „Zd dded d „Zdefd„ƒYZedkrejdddddddddddddddd d!d"d#d$d%d&d'd(d)d*d+d,d-d.d/d0g ƒZeed1d2ƒZdd3lmZmZeeed4d5d6eeƒd1d7ƒƒZ ee ƒZ!ndS(8s+ Seasonal Decomposition by Moving Averages iÿÿÿÿ(tlmaptranget iteritemsN(tnanmeani(t_maybe_get_pandas_wrapper_freqt_maybe_get_pandas_wrapper(tconvolution_filter(tfreq_to_periodcCs?tjgt|ƒD]%}t||d|…ddƒ^qƒS(s£ Return means for each period in x. freq is an int that gives the number of periods per cycle. E.g., 12 for monthly. NaNs are ignored in the mean. Ntaxisi(tnptarrayRt pd_nanmean(txtfreqti((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pyt seasonal_mean sc Cs×td„t|ƒDƒƒ}|jddtd„t|ddd…ƒDƒƒ}t|||ƒ}t|||ƒ}tjjtjtj ||ƒtj ||ƒf|||!ddƒd\}}tj d|ƒtj|tj|j }|j dkr|j ƒ}n|||*tjjtjtj ||ƒtj ||ƒf|||!ddƒd\}}tj |d|jdƒtj|tj|j }|j dkrÅ|j ƒ}n|||d)|S(s™ Replace nan values on trend's end-points with least-squares extrapolated values with regression considering npoints closest defined points. css3|])\}}tjtj|ƒƒs|VqdS(N(R tanytisnan(t.0Rtvals((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pys s iicss3|])\}}tjtj|ƒƒs|VqdS(N(R RR(RRR((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pys s Niÿÿÿÿtrcond(tnextt enumeratetshapetmintmaxR tlinalgtlstsqtc_tarangetonestTtndimtsqueeze( ttrendtnpointstfronttbackt front_lastt back_firsttktntextra((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pyt_extrapolate_trends&8 ) +  ) 6tadditiveic Cs¿|dkr!t|ƒ\}}nt|ƒ}d}tj|ƒjƒ}t|ƒ}tjtj|ƒƒs{t dƒ‚n|j dƒr±tj |dkƒr±t dƒ‚q±n|dkrí|dk rÞt |ƒ}|}qít dƒ‚n|dkrQ|ddkr8tj dgdg|ddgƒ|}qQtjd ||ƒ}nt|ƒd} t||| ƒ} |d krŒ|d}n|dkr®t| |dƒ} n|j dƒrÊ|| } n || } t| |ƒ} |j dƒr| tj| d dƒ:} n| tj| d dƒ8} tj| j||dƒj| } |j dƒrk|| | }n | | }t|| | ||gƒ}td |dd |dd|dd|dƒS(sÝ Seasonal decomposition using moving averages Parameters ---------- x : array-like Time series. If 2d, individual series are in columns. model : str {"additive", "multiplicative"} Type of seasonal component. Abbreviations are accepted. filt : array-like The filter coefficients for filtering out the seasonal component. The concrete moving average method used in filtering is determined by two_sided. freq : int, optional Frequency of the series. Must be used if x is not a pandas object. Overrides default periodicity of x if x is a pandas object with a timeseries index. two_sided : bool The moving average method used in filtering. If True (default), a centered moving average is computed using the filt. If False, the filter coefficients are for past values only. extrapolate_trend : int or 'freq', optional If set to > 0, the trend resulting from the convolution is linear least-squares extrapolated on both ends (or the single one if two_sided is False) considering this many (+1) closest points. If set to 'freq', use `freq` closest points. Setting this parameter results in no NaN values in trend or resid components. Returns ------- results : obj A object with seasonal, trend, and resid attributes. Notes ----- This is a naive decomposition. More sophisticated methods should be preferred. The additive model is Y[t] = T[t] + S[t] + e[t] The multiplicative model is Y[t] = T[t] * S[t] * e[t] The seasonal component is first removed by applying a convolution filter to the data. The average of this smoothed series for each period is the returned seasonal component. See Also -------- statsmodels.tsa.filters.bk_filter.bkfilter statsmodels.tsa.filters.cf_filter.xffilter statsmodels.tsa.filters.hp_filter.hpfilter statsmodels.tsa.filters.convolution_filter s,This function does not handle missing valuestmisJMultiplicative seasonality is not appropriate for zero and negative valuesshYou must specify a freq or x must be a pandas object with a timeseries index with a freq not set to Noneigà?igð?R RtseasonalR"tresidtobservediN(tNoneRRR t asanyarrayR!tlentalltisfinitet ValueErrort startswithRRR trepeattintRR+RtmeanttileRRtDecomposeResult(R tmodeltfiltR t two_sidedtextrapolate_trendt_pandas_wrappertpfreqtnobstnsidesR"t detrendedtperiod_averagesR.R/tresults((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pytseasonal_decompose5sP6        /     $ R<cBseZd„Zd„ZRS(cKsCx*t|ƒD]\}}t|||ƒq Wt|jƒ|_dS(N(RtsetattrR3R0RC(tselftkwargstkeytvalue((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pyt__init__«scCsÎddlm}|ƒ}|jdddtƒ\}}t|jdƒr|jjd|dd tƒ|djd ƒ|j jd|dd tƒ|djd ƒ|j jd|d d tƒ|d jd ƒ|j jd|dd tƒ|djdƒn¼|dj|jƒ|djd ƒ|dj|j ƒ|djd ƒ|d j|j ƒ|d jd ƒ|dj|j ƒ|djdƒ|dj dƒ|dj d|jƒ|jƒ|S(Niÿÿÿÿ(t _import_mpliitsharextplottaxitlegendtObservedtTrenditSeasonalitResidualtTime(tstatsmodels.graphics.utilsROtsubplotstTruethasattrR0RQtFalset set_ylabelR"R.R/t set_xlabeltset_xlimRCt tight_layout(RJROtplttfigtaxes((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pyRQ°s0  (t__name__t __module__RNRQ(((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pyR<ªs t__main__iÎÿÿÿi¯i•iÖi÷iíiáiIiÙi)iiéiiÉiÃi°iQiïi€ifièi­iibi+isÿÿÿi³ÿÿÿióÿÿÿi}iiiÓÿÿÿi¸R i(t DataFramet date_rangetstarts1/1/1951tperiodstQ("t__doc__tstatsmodels.compat.pythonRRRtnumpyR tpandas.core.nanopsRR tfilters._utilsRRtfilters.filtertoolsRtstatsmodels.tsa.tsatoolsRRR+R1R[RHtobjectR<ReR R RGtpandasRhRiR3tdatatres(((s7lib/python2.7/site-packages/statsmodels/tsa/seasonal.pyts*    t# $-