M/Ph/RdZddlZddlmZGddZGddZdS) zp This module contains scatterplot smoothers, that is classes who generate a smooth fit of a set of (x,y) pairs. N)kernelsc>eZdZdZd dZdZdZdZdZdZ d Z dS) KernelSmootherz 1D Kernel Density Regression/Kernel Smoother Requires: x - array_like of x values y - array_like of y values Kernel - Kernel object, Default is Gaussian. Nc|tj}||_tj||_tj||_dSN)rGaussianKernelnparrayxy)selfr rr s k/var/www/html/test/jupyter/venv/lib/python3.11/site-packages/statsmodels/sandbox/nonparametric/smoothers.py__init__zKernelSmoother.__init__s> >%''F !!cdSrrs rfitzKernelSmoother.fits rcDtjfd|DS)Nc:g|]}|Srpredict.0xxrs r z+KernelSmoother.__call__..#s%666bb))666rr r rr s` r__call__zKernelSmoother.__call__"s(x6666A666777rctj|dkr&jjj|Stjfdtj|DS)z Returns the kernel smoothed prediction at x If x is a real number then a single value is returned. Otherwise an attempt is made to cast x to numpy.ndarray and an array of corresponding y-points is returned. rc\g|](}jjj|)Sr)r smoothr rrs rrz*KernelSmoother.predict..1sC@@@T[//CC@@@r)r sizer r$r rr r s` rrzKernelSmoother.predict%s 71::??;%%dfdfa88 88@@@@358A;;@@@AA Arct|trQtjj}||dd|}|}||fStjfd|DS)ae Returns the fitted curve and 1-sigma upper and lower point-wise confidence. These bounds are based on variance only, and do not include the bias. If the bandwidth is much larger than the curvature of the underlying function then the bias could be large. x is the points on which you want to evaluate the fit and the errors. Alternatively if x is specified as a positive integer, then the fit and confidence bands points will be returned after every xth sample point - so they are closer together where the data is denser. Nc\g|](}jjj|)Sr)r smoothconfr rrs rrz'KernelSmoother.conf..JsGMMMDF"[33DFDFBGGMMMr) isinstanceintr r r sortconf)rr sorted_xconfxconffits` rr,zKernelSmoother.conf4s a   Nx''H MMOOOSSqSMEii&&G7# #8MMMMJKMMMNN NrcDtjfd|DS)Nc\g|](}jjj|)Sr)r smoothvarr rrs rrz&KernelSmoother.var..Os1OOOr..tvtvrBBOOOrrr s` rvarzKernelSmoother.varNs(xOOOOQOOOPPPrcPtj||Sr)r sqrtr3r s rstdzKernelSmoother.stdQswtxx{{###rr) __name__ __module__ __qualname____doc__rrr!rr,r3r6rrrrrs   888 A A ANNN4QQQ$$$$$rrcReZdZdZd dZdZdZd dZdZdZ d d Z d d Z d d Z dS) PolySmootherz Polynomial smoother up to a given order. Fit based on weighted least squares. The x values can be specified at instantiation or when called. This is a 3 liner with OLS or WLS, see test. It's here as a test smoother for GAM Nc:||_tj|dzftj|_hjdkr!t djdddftjfdt|dzDj |_ dSdS)Nrz3Warning: 2d x detected in PolySmoother init, shape:rcg|]}|zSrrrir s rrz)PolySmoother.__init__..rs<<<q!t<<.s>>>Q1a4>>>r) rErFrGr r rHrArJrDsqueezedotrI)rr rJs ` rrzPolySmoother.predicts =vzzNPQPWXXXAAAaC&>>>>% 1 *=*=>>>??AAA 71:+ + +:bfQ 2233 3:bfQS$)4455 5rc|jd|_|jdkr |dddf}|&tj|rd}d}ntj|dddf} t|dstdn_jdkrtdjtj fdt|j dzDj |_|j|z}||z}tj||dd|_tj|j|_dS) NrrrJzx needed to fit PolySmootherrZcg|]}|zSrrr?s rrz$PolySmoother.fit..sCCC!1CCCr)rcond)rGrVrEr isnanallr5hasattr ValueErrorrFr rHrArIrJlinalglstsqrDr\params)rrr weights_wrJ_ys ` rrzPolySmoother.fitsH 6Q;;!!!D& A ?bhw//3355?GBB!!!!!D&)B 94%% A !?@@@ AvzzNPQPWXXXXCCCCuTZ\/B/BCCCDDFDF FRK VIOOArO44Q7 j++ rr)NN) r7r8r9r:rrMrLrQr$rWr!rrrrrr<r<Ts& @ @ @ @      '''''' !!!!6666",,,,,,rr<)r:numpyr rrr<rrrrnsC$C$C$C$C$C$C$C$Jk,k,k,k,k,k,k,k,k,k,r