泊松回归，属于地理学以及GIS空间分析常用模型。适合应用于因变量为计数型变量的实例。模型基本知识移步百度，以下为亲测实例代码。一定有错漏，欢迎交流~ # _*_ coding: utf-8 _*_import pandas as pdimport numpy as npimport statsmodels.api as smfrom statsmodels.formula.api import ols #加载ols模型from statsmodels.formula.api import poissonimport matplotlib.pyplot as pltdata = pd.read_csv(“C:\\变量st.csv”)print(data.head())y = data[‘工作日D’]x1 = data[‘X_NDVI’]x2 = data[‘X_街景绿化’]x3 = data[‘X_道路里程’]x4 = data[‘X_坡度’]x5 = data[‘X_公交站’]x6 = data[‘X_地铁站’]x7 = data[‘X_购物点’]x8 = data[‘X_混合’]x = np.column_stack((x1, x2, x3, x4, x5, x6, x7, x8))# possion回归model = sm.GLM(y,x,family=sm.families.Poisson())# model=poisson(y,x)results = model.fit()print(results.summary())
输出模型结果：
Generalized Linear Model Regression Results ==============================================================================Dep. Variable: 工作日D No. Observations: 43Model: GLM Df Residuals: 35Model Family: Poisson Df Model: 7Link Function: log Scale: 1.0000Method: IRLS Log-Likelihood: -2.4490e+05Date: Sat, 01 Aug 2020 Deviance: 4.8933e+05Time: 17:51:09 Pearson chi2: 9.94e+05No. Iterations: 7 Covariance Type: nonrobust ============================================================================== coef std err z P>|z| [0.025 0.975]——————————————————————————x1 -0.2752 0.014 -19.714 0.000 -0.303 -0.248×2 5.4745 0.010 523.050 0.000 5.454 5.495×3 6.6649 0.011 618.719 0.000 6.644 6.686×4 4.8497 0.017 282.112 0.000 4.816 4.883×5 0.9419 0.013 69.895 0.000 0.915 0.968×6 0.3792 0.010 36.683 0.000 0.359 0.399×7 0.2701 0.012 22.606 0.000 0.247 0.294×8 1.4817 0.008 194.155 0.000 1.467 1.497==============================================================================
