数据分析第六章实践

import numpy as np
import pandas as pd
inputfile = 'C:/Users/Lenore/Desktop/data/data.csv' # 输入的数据文件
data = pd.read_csv(inputfile) # 读取数据

# 描述性统计分析
description = [data.min(), data.max(), data.mean(), data.std()]  # 依次计算最小值、最大值、均值、标准差
description = pd.DataFrame(description, index = ['Min', 'Max', 'Mean', 'STD']).T  # 将结果存入数据框
print('描述性统计结果_3042：\n',np.round(description, 2))  # 保留两位小数

# 相关性分析
corr = data.corr(method = 'pearson')  # 计算相关系数矩阵
print('相关系数矩阵为_3042：\n',np.round(corr, 2))  # 保留两位小数

# 绘制热力图
import matplotlib.pyplot as plt
import seaborn as sns
plt.subplots(figsize=(10, 10)) # 设置画面大小 
sns.heatmap(corr, annot=True, vmax=1, square=True,cmap="crest") 
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
plt.title('相关性热力图_3042',fontsize=15)
plt.show()
plt.close

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

inputfile='C:/Users/Lenore/Desktop/data/data.csv'
data=pd.read_csv(inputfile)
lasso=Lasso(1000)
lasso.fit(data.iloc[:,0:13],data['y'])
print('相关系数为_3042:',np.round(lasso.coef_,5))
print('相关系数非零个数为_3042:',np.sum(lasso.coef_!=0))
mask=lasso.coef_!=0
print('相关系数是否为零_3042:',mask)
mask=np.append(mask,True)

outputfile='C:/Users/Lenore/Desktop/data/new_reg_data.csv'
new_reg_data=data.iloc[:,mask]
new_reg_data.to_csv(outputfile)
print('输出数据的维度为_3042:',new_reg_data.shape)

import sys
sys.path.append('C:/Users/Lenore/Desktop/data')  # 设置路径
import numpy as np
import pandas as pd
from GM11 import GM11  # 引入自编的灰色预测函数

inputfile1 = 'C:/Users/Lenore/Desktop/data/new_reg_data.csv'  # 输入的数据文件
inputfile2 = 'C:/Users/Lenore/Desktop/data/data.csv'  # 输入的数据文件
new_reg_data = pd.read_csv(inputfile1)  # 读取经过特征选择后的数据
data = pd.read_csv(inputfile2)  # 读取总的数据
new_reg_data.index = range(1994, 2014)
new_reg_data.loc[2014] = None
new_reg_data.loc[2015] = None
new_reg_data.loc[2016] = None
l = ['x1', 'x3', 'x4', 'x5', 'x6', 'x7', 'x8', 'x13']
for i in l:
  f = GM11(new_reg_data.loc[range(1994, 2014),i].to_numpy())[0]
  new_reg_data.loc[2014,i] = f(len(new_reg_data)-2)  # 2014年预测结果
  new_reg_data.loc[2015,i] = f(len(new_reg_data)-1)  # 2015年预测结果
  new_reg_data.loc[2016,i] = f(len(new_reg_data))  # 2015年预测结果
  new_reg_data[i] = new_reg_data[i].round(2)  # 保留两位小数
outputfile = 'C:/Users/Lenore/Desktop/data/new_reg_data_GM21.xls'  # 灰色预测后保存的路径
y = list(data['y'].values)  # 提取财政收入列，合并至新数据框中
y.extend([np.nan,np.nan,np.nan])
new_reg_data['y'] = y
new_reg_data.to_excel(outputfile)  # 结果输出
print('预测结果为_3042：\n',new_reg_data.loc[2014:2016,:])  # 预测结果展示

import matplotlib.pyplot as plt
from sklearn.svm import LinearSVR

inputfile = 'C:/Users/Lenore/Desktop/data/new_reg_data_GM21.xls'  # 灰色预测后保存的路径
data = pd.read_excel(inputfile)  # 读取数据
#构建支持向量回归预测模型
feature = ['x1', 'x3', 'x4', 'x5', 'x6', 'x7', 'x8', 'x13']  # 属性所在列
data_train = new_reg_data.loc[range(1994,2014)].copy()  # 取2014年前的数据建模
data_mean = data_train.mean()
data_std = data_train.std()
data_train = (data_train - data_mean)/data_std  # 数据标准化
x_train = data_train[feature].to_numpy() # 属性数据
y_train = data_train['y'].to_numpy()  # 标签数据

model = LinearSVR()  # 调用LinearSVR()函数
model.fit(x_train,y_train)
x = ((new_reg_data[feature] - data_mean[feature])/data_std[feature]).to_numpy()  # 预测，并还原结果。
new_reg_data['y_pred'] = model.predict(x) * data_std['y'] + data_mean['y']
outputfile = 'C:/Users/Lenore/Desktop/data/new_reg_data_GM21_revenue.xls'  # SVR预测后保存的结果
data.to_excel(outputfile)
print('真实值与预测值分别为_3042:\n',new_reg_data[['y','y_pred']])

fig = new_reg_data[['y','y_pred']].plot(style=['b-o','r-*'])  # 画出预测结果图
plt.title('2014-2016地方财政收入真实值与预测值对比图_3042')
plt.show()