Python_简单交互作用回归分析及散点图矩阵

2019-04-15 14:11发布

站内文章 / PIC单片机
14300 0 960

1、数据准备

import urllib.request import matplotlib.pyplot as plt plt.style.use('ggplot') import pandas as pd # 下载同时显示进度条 def jdt(down, pic, all): """ @down : 已经下载的数据块 @pic : 数据块的大小 @all : 远程文件的大小 """ jd = 100.0 * down * pic / all if jd > 100: jd = 100 print('%.2f%%'%jd) url = 'http://lib.stat.cmu.edu/datasets/CPS_85_Wages' fil_out = r'D:Python_datawages.txt' urllib.request.urlretrieve(url, fil_out,jdt)

2、数据读取及简单探索

查看下载文件，前27行为无效数据, 后6行也是无效
数据描述如下：
EDUCATION: Number of years of education.
SOUTH: Indicator variable for Southern Region (1=Person lives in
South, 0=Person lives elsewhere).
SEX: Indicator variable for sex (1=Female, 0=Male).
EXPERIENCE: Number of years of work experience.
UNION: Indicator variable for union membership (1=Union member,
0=Not union member).
WAGE: Wage (dollars per hour).
AGE: Age (years).
RACE: Race (1=Other, 2=Hispanic, 3=White).
OCCUPATION: Occupational category (1=Management, 2=Sales, 3=Clerical, 4=Service, 5=Professional, 6=Other).
SECTOR: Sector (0=Other, 1=Manufacturing, 2=Construction).
MARR: Marital Status (0=Unmarried, 1=Married)
name = ['EDUCATION', 'SOUTH', 'SEX', 'EXPERIENCE', 'UNION', 'WAGE', 'AGE', 'RACE', 'OCCUPATION', 'SECTOR', 'MARR'] data_wage = pd.read_table(fil_out, skiprows=27, skipfooter = 6, sep = ' ', header=None ) data_wage.columns = name # 去掉大部分分类变量看相关 d = ['EDUCATION', 'SEX', 'EXPERIENCE', 'WAGE', 'AGE'] pd.plotting.scatter_matrix(data_wage[d], color='steelblue') plt.show() 在这里插入图片描述

3、绘制交互作用相关性散点

# 绘制回归相关性矩阵+交互作用 import seaborn seaborn.pairplot(data_wage, vars=['WAGE', 'AGE', 'EDUCATION'] ,kind = 'reg', hue = 'SEX', size=5) plt.show() 在这里插入图片描述

4、建模分析

# 单变量回归绘图+交互作用 seaborn.lmplot(y='WAGE', x = 'EDUCATION', hue = 'SEX',data=data_wage) plt.show() # 建模 from statsmodels.formula.api import ols results = ols('WAGE~EDUCATION + C(SEX) + EDUCATION*C(SEX)', data = data_wage).fit() print(results.summary()) 结果如下：
在这里插入图片描述 ============================================================================== Dep. Variable: WAGE R-squared: 0.190 Model: OLS Adj. R-squared: 0.186 Method: Least Squares F-statistic: 41.50 Date: Sun, 30 Sep 2018 Prob (F-statistic): 4.24e-24 Time: 18:07:55 Log-Likelihood: -1575.0 No. Observations: 534 AIC: 3158. Df Residuals: 530 BIC: 3175. Df Model: 3 Covariance Type: nonrobust ========================================================================================= coef std err t P>|t| [0.025 0.975] ----------------------------------------------------------------------------------------- Intercept 1.1046 1.314 0.841 0.401 -1.476 3.685 C(SEX)[T.1] -4.3704 2.085 -2.096 0.037 -8.466 -0.274 EDUCATION 0.6831 0.099 6.918 0.000 0.489 0.877 EDUCATION:C(SEX)[T.1] 0.1725 0.157 1.098 0.273 -0.136 0.481 ============================================================================== Omnibus: 208.151 Durbin-Watson: 1.863 Prob(Omnibus): 0.000 Jarque-Bera (JB): 1278.081 Skew: 1.587 Prob(JB): 2.94e-278 Kurtosis: 9.883 Cond. No. 170. ==============================================================================
就该数据而言，男性受教育作用要比女性受教育作用大，这可能是女性存在生育期工作空挡和社会对男性偏向造成的

热门文章