在python中指定线性回归

如何解决在python中指定线性回归

我尝试对线性回归实施可靠的标准误差，但收到以下错误： LinAlgError: Singular matrix

以下是引发错误的代码：

#Get robust standard errors
Robust=sm.OLS(y,X,M=sm.robust.norms.HuberT()).fit()
Robust.summary()
qq = sm.qqplot(Robust.resid,line = 'r') 
Robust2=sm.RLM(y,M=sm.robust.norms.HuberT()).fit(cov="H2")
Robust2.summary()
shapiro(residual)

pred_val2 = Robust2.fittedvalues.copy()
true_val = regression_data_performance['CRSP_return_p1'].values.copy()
residual2 = true_val - pred_val2
shapiro(residual2)


data=['Sales','up_revision','pos_EPS','filing_range','time','DebttoAssets','ShareOverhang','VentureBacked','CRSP_ret_SP500_prior30days','CRSP_vola_SP500_prior30days','top_tier_dummy','ProceedsAmtInThisMktMil','Age','MainSICCode_industry48_y_Soda','MainSICCode_industry48_y_PerSv','MainSICCode_industry48_y_Rtail','MainSICCode_industry48_y_Rubbr','MainSICCode_industry48_y_Ships','MainSICCode_industry48_y_Smoke','MainSICCode_industry48_y_Steel','MainSICCode_industry48_y_Toys','MainSICCode_industry48_y_Txtls','MainSICCode_industry48_y_Whlsl','CRSP_return_p1']


regression_data_performance=regression_data_performance[~isNaN(regression_data_performance['filing_range'])]
data=regression_data_performance[data]
data1=data.fillna(0)
data2 = preprocessing.scale(data1)

如果有帮助，请参见完整的错误消息：

LinAlgError                               Traceback (most recent call last)
<ipython-input-33-1febd940375c> in <module>
      3 Robust.summary()
      4 qq = sm.qqplot(Robust.resid,line = 'r')
----> 5 Robust2=sm.RLM(y,M=sm.robust.norms.HuberT()).fit(cov="H2")
      6 Robust2.summary()
      7 shapiro(residual)

~\anaconda3\lib\site-packages\statsmodels\robust\robust_linear_model.py in fit(self,maxiter,tol,scale_est,init,cov,update_scale,conv,start_params)
    299             converged = _check_convergence(criterion,iteration,maxiter)
    300         results = RLMResults(self,wls_results.params,--> 301                              self.normalized_cov_params,self.scale)
    302 
    303         history['iteration'] = iteration

~\anaconda3\lib\site-packages\statsmodels\robust\robust_linear_model.py in __init__(self,model,params,normalized_cov_params,scale)
    410         self.data_in_cache = ['sresid']
    411 
--> 412         self.cov_params_default = self.bcov_scaled
    413         # TODO: "pvals" should come from chisq on bse?
    414 

pandas\_libs\properties.pyx in pandas._libs.properties.CachedProperty.__get__()

~\anaconda3\lib\site-packages\statsmodels\robust\robust_linear_model.py in bcov_scaled(self)
    453             W = np.dot(model.M.psi_deriv(self.sresid) * model.exog.T,454                        model.exog)
--> 455             W_inv = np.linalg.inv(W)
    456             # [W_jk]^-1 = [SUM(psi_deriv(Sr_i)*x_ij*x_jk)]^-1
    457             # where Sr are the standardized residuals

<__array_function__ internals> in inv(*args,**kwargs)

~\anaconda3\lib\site-packages\numpy\linalg\linalg.py in inv(a)
    545     signature = 'D->D' if isComplexType(t) else 'd->d'
    546     extobj = get_linalg_error_extobj(_raise_linalgerror_singular)
--> 547     ainv = _umath_linalg.inv(a,signature=signature,extobj=extobj)
    548     return wrap(ainv.astype(result_t,copy=False))
    549 

~\anaconda3\lib\site-packages\numpy\linalg\linalg.py in _raise_linalgerror_singular(err,flag)
     95 
     96 def _raise_linalgerror_singular(err,flag):
---> 97     raise LinAlgError("Singular matrix")
     98 
     99 def _raise_linalgerror_nonposdef(err,flag):

LinAlgError: Singular matrix

有人知道我该如何解决错误？如果您需要更多信息，请告诉我。