ó áp7]c@sCdZddlZdded„Zded„ZddgZdS(s„Principal Component Analysis Created on Tue Sep 29 20:11:23 2009 Author: josef-pktd TODO : add class for better reuse of results i˙˙˙˙Nic CsUtj|ƒ}|r'|jdƒ}ntj|jdƒ}||8}tj|ddƒ}tjj|ƒ\}}tj|ƒ} | ddd…} |dd…| f}|| }|dkrţ||jdkrţ|dd…d|…f}|| }n|r|tj |ƒ}ntj ||ƒ} tj | |j ƒ|} | | ||fS(sáprincipal components with eigenvector decomposition similar to princomp in matlab Parameters ---------- data : ndarray, 2d data with observations by rows and variables in columns keepdim : integer number of eigenvectors to keep if keepdim is zero, then all eigenvectors are included normalize : boolean if true, then eigenvectors are normalized by sqrt of eigenvalues demean : boolean if true, then the column mean is subtracted from the data Returns ------- xreduced : ndarray, 2d, (nobs, nvars) projection of the data x on the kept eigenvectors factors : ndarray, 2d, (nobs, nfactors) factor matrix, given by np.dot(x, evecs) evals : ndarray, 2d, (nobs, nfactors) eigenvalues evecs : ndarray, 2d, (nobs, nfactors) eigenvectors, normalized if normalize is true Notes ----- See Also -------- pcasvd : principal component analysis using svd iitrowvarNi˙˙˙˙( tnptarraytmeantzerostshapetcovtlinalgteigtargsorttsqrttdottT( tdatatkeepdimt normalizetdemeantxtmtxcovtevalstevecstindicestfactorstxreduced((sBlib/python2.7/site-packages/statsmodels/sandbox/tools/tools_pca.pytpcas&#   c CsB|j\}}tj|ƒ}|r6|jdƒ}nd}||8}tjj|jddƒ\}}} tj|j|jƒj} |rÓtj| dd…d|…f|dd…d|…fjƒ|} n|} |}d|f|d|jdd} | | dd…d|…f| | |dd…d|…ffS(s’principal components with svd Parameters ---------- data : ndarray, 2d data with observations by rows and variables in columns keepdim : integer number of eigenvectors to keep if keepdim is zero, then all eigenvectors are included demean : boolean if true, then the column mean is subtracted from the data Returns ------- xreduced : ndarray, 2d, (nobs, nvars) projection of the data x on the kept eigenvectors factors : ndarray, 2d, (nobs, nfactors) factor matrix, given by np.dot(x, evecs) evals : ndarray, 2d, (nobs, nfactors) eigenvalues evecs : ndarray, 2d, (nobs, nfactors) eigenvectors, normalized if normalize is true See Also -------- pca : principal component analysis using eigenvector decomposition Notes ----- This doesn't have yet the normalize option of pca. it full_matricesiNs print reassigning keepdim to maxi(RRRRRtsvdR R ( R RRtnobstnvarsRRtUtstvRRR((sBlib/python2.7/site-packages/statsmodels/sandbox/tools/tools_pca.pytpcasvdWs! $H RR!(t__doc__tnumpyRtTrueRR!t__all__(((sBlib/python2.7/site-packages/statsmodels/sandbox/tools/tools_pca.pyt s I=