财政收入影响因素分析及预测

Lasso 回归选取关键属性

#%%
import numpy as np
import pandas as pd
from sklearn.linear_model import Lasso

inputfile = './data/data.csv'
data = pd.read_csv(inputfile)
lasso = Lasso(1000)
lasso.fit(data.iloc[:, 0:14], data['y'])
print('相关系数为 :', np.round(lasso.coef_, 5))

#%%
print('相关系数非零个数为:', np.sum(lasso.coef_ != 0))

#%%
mask = lasso.coef_ != 0
print('相关系数是否为零:', mask)
#%%
outputfile = './data/new_reg_data.csv'
new_reg_data = data.iloc[:]
print(new_reg_data)
new_reg_data.to_csv(outputfile)
print('输出数据的维度为:', new_reg_data.shape)

二、构建灰色预测模型并预测

#%%
import numpy as np
import pandas as pd
from data.GM11 import GM11
#%%
inputfile1 = './data/new_reg_data.csv'
inputfile2 = './data/data.csv'
new_reg_data = pd.read_csv(inputfile1)
data = pd.read_csv(inputfile2)
new_reg_data.index = range(1997,2017)
new_reg_data.loc[2017] = None
new_reg_data.loc[2018] = None
cols = ['x1','x3','x4','x5','x6','x7','x8','x13']
for i in cols:
    f = GM11(new_reg_data.loc[range(1997,2016),i].values)[0]
    new_reg_data.loc[2017,i] = f(len(new_reg_data)-1)
    new_reg_data.loc[2018,i] = f(len(new_reg_data))
    new_reg_data[i] = new_reg_data[i].round(2)
outputfile = './data/new_reg_data_GM11.xls'
y =list(data['y'].values)
y.extend([np.nan,np.nan])
new_reg_data['y'] = y

new_reg_data.to_excel(outputfile)
print('预测结果为:\n',new_reg_data.loc[2014:2016,:])

三、构建支持向量回归预测模型

#%%
import matplotlib.pyplot as plt
from sklearn.svm import LinearSVR
import pandas as pd
import numpy as np
#%%
inputfile = './data/new_reg_data_GM11.xls'
data = pd.read_excel(inputfile)
data = data.replace(np.nan,0)
feature = ['x1','x3','x4','x5','x6','x7','x8','x13']
data.index = range(1997,2019)
data_train = data.loc[range(1997,2019)].copy()
# print(np.isnan(data).any())
data_mean = data_train.mean()
data_std = data_train.std()
data_train = (data_train - data_mean) / data_std
x_train = data_train[feature].values
y_train = data_train['y'].values
#%%
linearsvr = LinearSVR()
linearsvr.fit(x_train,y_train)
x = ((data[feature] -  data_mean[feature])/data_std[feature]).values
#%%
data[u'y_pred'] = linearsvr.predict(x) * data_std['y'] +data_mean['y']
outputfile = './data/new_reg_data_GM11_revenue.xls'
data.to_excel(outputfile)
#%%
print('真实值与预测值分别为:\n',data[['y','y_pred']])
#%%
data = data.loc[1997:2016,:]
print(data)
fig = data[['y','y_pred']].plot(style = ['b-o','y-*'])

四、结果