在pycharm中ols拟合出现nan,inf

Question

在预测碳排放量时，我使用ols模型，结果出现很多nan,inf

代码

import numpy as np
import pandas as pd
#载入数据
excel_path = r"Testingdata_shiyou.xlsx"
data_df = pd.read_excel(excel_path)
# 给列名重新命名，每一列就是一个自（因）变量的观测值
data_df.columns = ['year', 'I', 'P', 'A', 'Te', 'Ts', 'Td', 'Tc']
# 量纲恢复

# 数据标准化
def normalize(data):
    """对因变量作标准化变换，标准化因子是标准差"""
    return (data - data.mean()) / data.std()
#%%
import statsmodels.api as sm
X = np.log(data_df.iloc[:, 2:])
def mask_nan(X,value):
    '''
    用value值来代替nan 或inf
    '''
    X_values= tf.add(tf.zeros_like(X),value)
    mask = tf.math.is_finite(X)
    y = tf.where(mask,X,X_values)
    return y

y = np.log(data_df.iloc[:, 1])
X = sm.add_constant(X)
ols_model = sm.OLS(y,X).fit()
print(ols_model.summary())

结果

                            OLS Regression Results                            
==============================================================================
Dep. Variable:                      I   R-squared:                       1.000
Model:                            OLS   Adj. R-squared:                    nan
Method:                 Least Squares   F-statistic:                       nan
Date:                Wed, 13 Apr 2022   Prob (F-statistic):                nan
Time:                        16:29:49   Log-Likelihood:                 148.10
No. Observations:                   5   AIC:                            -286.2
Df Residuals:                       0   BIC:                            -288.2
Df Model:                           4                                         
Covariance Type:            nonrobust                                         
==============================================================================
                 coef    std err          t      P>|t|      [0.025      0.975]
------------------------------------------------------------------------------
const         -0.0678        inf         -0        nan         nan         nan
P              0.9989        inf          0        nan         nan         nan
A              0.9892        inf          0        nan         nan         nan
Te             0.9672        inf          0        nan         nan         nan
Ts            -0.2405        inf         -0        nan         nan         nan
Td             0.0003        inf          0        nan         nan         nan
Tc             0.9808        inf          0        nan         nan         nan
==============================================================================
Omnibus:                          nan   Durbin-Watson:                   0.035
Prob(Omnibus):                    nan   Jarque-Bera (JB):                0.362
Skew:                          -0.201   Prob(JB):                        0.834
Kurtosis:                       1.745   Cond. No.                         631.
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The input rank is higher than the number of observations.

我有试过这位的定义mask函数，但结果没变化https://blog.csdn.net/qq_25737169/article/details/78847691?spm=1001.2101.3001.6650.5&utm_medium=distribute.pc_relevant.none-task-blog-2%7Edefault%7EBlogCommendFromBaidu%7ERate-5.pc_relevant_default&depth_1-utm_source=distribute.pc_relevant.none-task-blog-2%7Edefault%7EBlogCommendFromBaidu%7ERate-5.pc_relevant_default&utm_relevant_index=10

想请各位帮忙看看如何解决结果里的nan和inf

Answer 1

NaN：not a number
不是一个数字
每个变量都Print一下“变量名称”+变量值，看看哪个变量不是数字，这样做还可以看出程序执行到哪步，哪里有问题。
这么几个变量，很快就能排查出问题。
如果用这个方法解决了问题，望采纳。