M/Ph"|dZddlZddlmZddlmZddlmZm Z ddl m Z GddZ Gd d e Z dS) ay This module implements empirical likelihood regression that is forced through the origin. This is different than regression not forced through the origin because the maximum empirical likelihood estimate is calculated with a vector of ones in the exogenous matrix but restricts the intercept parameter to be 0. This results in significantly more narrow confidence intervals and different parameter estimates. For notes on regression not forced through the origin, see empirical likelihood methods in the OLSResults class. General References ------------------ Owen, A.B. (2001). Empirical Likelihood. Chapman and Hall. p. 82. N)optimize)chi2)OLSRegressionResults) add_constantc&eZdZdZdZdZddZdS)ELOriginRegressa/ Empirical Likelihood inference and estimation for linear regression through the origin. Parameters ---------- endog: nx1 array Array of response variables. exog: nxk array Array of exogenous variables. Assumes no array of ones Attributes ---------- endog : nx1 array Array of response variables exog : nxk array Array of exogenous variables. Assumes no array of ones. nobs : float Number of observations. nvar : float Number of exogenous regressors. c||_||_|jjd|_ t |jd|_dS#t $r d|_YdSwxYw)Nrg?)endogexogshapenobsfloatnvar IndexError)selfr r s a/var/www/html/test/jupyter/venv/lib/python3.11/site-packages/statsmodels/emplike/originregress.py__init__zELOriginRegress.__init__7sf  IOA&  djm,,DIII   DIIII sAAAct|jd}t|j|}|}|t jdgt jdgd}t j|d}|d}t j t j |d}t||||S)z Fits the model and provides regression results. Returns ------- Results : class Empirical likelihood regression class. Tprependrr ) ret_params) rr rr fitel_testnparraysqueezesumlog OriginResults)r exog_withrestricted_modelrestricted_fit restricted_elparams beta_hat_llrllfs rrzELOriginRegress.fit@s!D999 tz955)--//&.. ! rx}}/44 M!,--$Q' fRVM!,--..-v|SIIINc\||j}tjt|d|S)NTr)r rdotr)rr(r s rpredictzELOriginRegress.predictSs- <9Dvl4666???r+N)__name__ __module__ __qualname____doc__rrr.r+rr r sY4JJJ&@@@@@@r+r c2eZdZdZdZ d dZ d d ZdS) r#a A Results class for empirical likelihood regression through the origin. Parameters ---------- model : class An OLS model with an intercept. params : 1darray Fitted parameters. est_llr : float The log likelihood ratio of the model with the intercept restricted to 0 at the maximum likelihood estimates of the parameters. llr_restricted/llr_unrestricted llf_el : float The log likelihood of the fitted model with the intercept restricted to 0. Attributes ---------- model : class An OLS model with an intercept. params : 1darray Fitted parameter. llr : float The log likelihood ratio of the maximum empirical likelihood estimate. llf_el : float The log likelihood of the fitted model with the intercept restricted to 0. Notes ----- IMPORTANT. Since EL estimation does not drop the intercept parameter but instead estimates the slope parameters conditional on the slope parameter being 0, the first element for params will be the intercept, which is restricted to 0. IMPORTANT. This class inherits from RegressionResults but inference is conducted via empirical likelihood. Therefore, any methods that require an estimate of the covariance matrix will not function. Instead use el_test and conf_int_el to conduct inference. Examples -------- >>> import statsmodels.api as sm >>> data = sm.datasets.bc.load() >>> model = sm.emplike.ELOriginRegress(data.endog, data.exog) >>> fitted = model.fit() >>> fitted.params # 0 is the intercept term. array([ 0. , 0.00351813]) >>> fitted.el_test(np.array([.0034]), np.array([1])) (3.6696503297979302, 0.055411808127497755) >>> fitted.conf_int_el(1) (0.0033971871114706867, 0.0036373150174892847) # No covariance matrix so normal inference is not valid >>> fitted.conf_int() TypeError: unsupported operand type(s) for *: 'instancemethod' and 'float' cb||_tj||_||_||_dSr/)modelrr r(llrllf_el)rr7r(est_llrr9s rrzOriginResults.__init__s, j((  r+nmr rc\tjd|f}tjd|f}|j|||||}|d}||jz }t j||jjj ddz } |r || |dfS|| fS)a Returns the llr and p-value for a hypothesized parameter value for a regression that goes through the origin. Parameters ---------- b0_vals : 1darray The hypothesized value to be tested. param_num : 1darray Which parameters to test. Note this uses python indexing but the '0' parameter refers to the intercept term, which is assumed 0. Therefore, param_num should be > 0. return_weights : bool If true, returns the weights that optimize the likelihood ratio at b0_vals. Default is False. method : str Can either be 'nm' for Nelder-Mead or 'powell' for Powell. The optimization method that optimizes over nuisance parameters. Default is 'nm'. stochastic_exog : bool When TRUE, the exogenous variables are assumed to be stochastic. When the regressors are nonstochastic, moment conditions are placed on the exogenous variables. Confidence intervals for stochastic regressors are at least as large as non-stochastic regressors. Default is TRUE. Returns ------- res : tuple pvalue and likelihood ratio. r)methodstochastic_exogreturn_weightsr r) rhstackr7rrr8rsfr r) rb0_vals param_numsr=r>r?test_resllr_testllr_respvals rrzOriginResults.el_testsJ)QL))Y:// :>>##++GZ2A1?,AAA;TX%ww  5a 81 <==  !D(1+- -D= r+N皙?Tc tjd|z d tjg|^tjjd}tj|d}|^tjjd}tj|d} fd}tjj } tj |tj|| } tj || tj|} | | fS)au Returns the confidence interval for a regression parameter when the regression is forced through the origin. Parameters ---------- param_num : int The parameter number to be tested. Note this uses python indexing but the '0' parameter refers to the intercept term. upper_bound : float The maximum value the upper confidence limit can be. The closer this is to the confidence limit, the quicker the computation. Default is .00001 confidence limit under normality. lower_bound : float The minimum value the lower confidence limit can be. Default is .00001 confidence limit under normality. sig : float, optional The significance level. Default .05. method : str, optional Algorithm to optimize of nuisance params. Can be 'nm' or 'powell'. Default is 'nm'. stochastic_exog : bool Default is True. Returns ------- ci: tuple The confidence interval for the parameter 'param_num'. r Ng-C6?rcvtj|g}|}|dz S)N)r=r>r)rrr)b0valr= param_numr0rr>s rfz$OriginResults.conf_int_el..fsC2$B,,IfoCq6B; r+) rppfrrasarrayr7rconf_intr r(rbrentq) rrM upper_bound lower_boundsigr=r>cirO_paramlowerlupperlrNs `` `` @r conf_int_elzOriginResults.conf_int_elsH@Xa#gq ! !Hi[))  DJNN,,55e<<==B:bm44Q7K  DJNN,,55e<<==B:bm44Q7K         DK 233BJ{$;$;VDDFBJ{,C,CDDr+)r;r r)NNrHr;T)r0r1r2r3rrr[r4r+rr#r#Ysj>>~ 37>?0!0!0!0!d269='+3 3 3 3 3 3 r+r#)r3numpyrscipyr scipy.statsr#statsmodels.regression.linear_modelrrstatsmodels.tools.toolsrr r#r4r+rras$FFFFFFFF000000:@:@:@:@:@:@:@:@zj j j j j %j j j j j r+