Python多元线性回归

# -*- coding: utf-8 -*-
"""
Created on Fri Aug 16 09:14:55 2019

@author: 6
"""
# import eddy_shape
import pickle
import glob
import numpy as np
import pdb
import datetime
import matplotlib.pyplot as plt
#chl=np.zeros((2937,200))
import h5py
import pdb
import json
import glob
import pickle
import calendar,os
import pdb
import scipy.io as sio
from scipy import signal
import  copy
from math import atan,pi,sin,cos
from matplotlib.path import Path
import math
import netCDF4
import gc
    
import numpy as np
import pandas as pd
from sklearn.datasets import load_boston


# # 读取网络数据
# boston = load_boston()
# # 数据包含14个字段，boston.data是前13个字段数据，boston.target是第13个字段'medv'的数据
# col = ['crim','zn','indus','chas','nox','rm','age','dis','rad','tax','ptratio','b','lstat']
# bostondf = pd.DataFrame(boston.data,columns=col)
# bostondf['medv']=boston.target  #待回归的变量结果 numpy
# bostondf.head()

# a=boston.data
# b=pd.DataFrame(a,columns=col)
# b['medv']=boston.target

# import statsmodels.formula.api as smf
# mod = smf.ols(formula='medv~crim+zn+indus+chas+nox+rm+age+dis+rad+tax+ptratio+b+lstat',data=b)
# res = mod.fit()
# print(res.summary())

'''test'''

b=np.random.randint(1,500,200)
c=np.random.randint(1,500,200)
a=b*2+c*4

col=['x','y']
d=np.array((b,c)).T

import statsmodels.formula.api as smf
bostondf=pd.DataFrame(d,columns=col)
bostondf['medv']=a
mod = smf.ols(formula='medv~x+y',data=bostondf)
res = mod.fit()
print(res.summary())

'''
coef代表回归系数
std 标准差
t检验
p>t 显著性检验
'''

#参考网站 https://blog.csdn.net/m0_53653974/article/details/124858229