ó áp7]c@s"dZddlZed„ZdS(sœCollection of alternative implementations for time series analysis >>> signal.fftconvolve(x,x[::-1])[len(x)-1:len(x)+10]/x.shape[0] array([ 2.12286549e+00, 1.27450889e+00, 7.86898619e-02, -5.80017553e-01, -5.74814915e-01, -2.28006995e-01, 9.39554926e-02, 2.00610244e-01, 1.32239575e-01, 1.24504352e-03, -8.81846018e-02]) >>> sm.tsa.stattools.acovf(X, fft=True)[:order+1] array([ 2.12286549e+00, 1.27450889e+00, 7.86898619e-02, -5.80017553e-01, -5.74814915e-01, -2.28006995e-01, 9.39554926e-02, 2.00610244e-01, 1.32239575e-01, 1.24504352e-03, -8.81846018e-02]) >>> import nitime.utils as ut >>> ut.autocov(s)[:order+1] array([ 2.12286549e+00, 1.27450889e+00, 7.86898619e-02, -5.80017553e-01, -5.74814915e-01, -2.28006995e-01, 9.39554926e-02, 2.00610244e-01, 1.32239575e-01, 1.24504352e-03, -8.81846018e-02]) iÿÿÿÿNcCsddlm}tj|ƒ}|r8||jƒ}n|j||ddd…ƒt|ƒdt|ƒd!|jddS(s`autocovariance function with call to fftconvolve, biased Parameters ---------- x : array_like timeseries, signal demean : boolean If true, then demean time series Returns ------- acovf : array autocovariance for data, same length as x might work for nd in parallel with time along axis 0 iÿÿÿÿ(tsignalNii i(tscipyRtnptasarraytmeant fftconvolvetlentshape(txtdemeanR((s>lib/python2.7/site-packages/statsmodels/sandbox/archive/tsa.pyt acovf_ffts (t__doc__tnumpyRtTrueR (((s>lib/python2.7/site-packages/statsmodels/sandbox/archive/tsa.pyts