python数据挖掘—绘制热力图，相关系数分析，基于GM11和ARIMA的财政收入预测

01绘制热力图

import numpy as np
import pandas as pd

inputfile = 'D://CourseAssignment//AI//DataPredict//data.csv'
data = pd.read_csv(inputfile)


#describe statistical analysis
description = [data.min(), data.max(), data.mean(), data.std()]
#calculate the min, max, mean, std
description = pd.DataFrame(description, index = ['Min', 'Max', 'Mean', 'STD']).T
#keep two decimals
print('describe the result of statisticl analysis : \n', np.round(description, 2))


#correlation analysis
corr = data.corr(method = 'pearson')
print('correlation coefficient martix: \n', np.round(corr, 2))


#draw thermodynamic diagram
import matplotlib.pyplot as plt
import seaborn as sns
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.subplots(figsize=(10, 10)) # set the size of canvas 
sns.heatmap(corr, annot=True, vmax=1, square=True, cmap="Reds")
plt.title('correlation thermodynamic diagram')
plt.show()
plt.close

 

 

 

 

 

 

02相关系数

from sklearn.linear_model import Lasso
lasso = Lasso(1000)   #use function Lasso, set 1000
lasso.fit(data.iloc[:,0:13], data['y'])

print('相关系数为: ', np.round(lasso.coef_, 5)) #output result
print('相关系数非0个数为：',np.sum(lasso.coef_ != 0))

mask = lasso.coef_ != 0;
print('相关系数是否为0：', mask)
mask = np.append(mask, True)
print('相关系数是否为0：', mask)

outputFile = r'D:\CourseAssignment\AI\new_reg_data3.csv'  #输出数据文件
new_reg_data = data.iloc[:, mask]                       #返回相关系数非零的？？
new_reg_data.to_csv(outputFile)                         #存储数据
print('输出数据的维度为：', new_reg_data.shape)                 #查看数据

 

 

 

 

03基于GM11的财政预测

#-*- coding: utf-8 -*-

import sys
sys.path.append('D://CourseAssignment//AI//DataPredict//code')  # 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 = r'D://CourseAssignment//AI//DataPredict//new_reg_data.csv'  # 输入的数据文件
inputfile2 = r'D://CourseAssignment//AI//DataPredict//data.csv'  # 输入的数据文件
new_reg_data = pd.read_csv(inputfile1, index_col = 0,header =0)  # 读取经过特征选择后的数据
data = pd.read_csv(inputfile2, header=0)  # 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].values)[0]
  new_reg_data.loc[2014,i] = f(len(new_reg_data)-1)  # 2014年预测结果
  new_reg_data.loc[2015,i] = f(len(new_reg_data))  # 2015年预测结果
  new_reg_data.loc[2016, i] = f(len(new_reg_data)+1)  # 2016年预测结果
  new_reg_data[i] = new_reg_data[i].round(2)  # 保留两位小数


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
outputfile = 'D://CourseAssignment//AI//DataPredict//new_reg_data_GM11.xls'  # 灰色预测后保存的路径
new_reg_data.to_excel(outputfile)  # 结果输出
print('预测结果为：\n',new_reg_data.loc[2014:2016,])  # 预测结果展示


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

inputfile = 'D://CourseAssignment//AI//DataPredict//new_reg_data_GM11.xls'  # 灰色预测后保存的路径
data = pd.read_excel(inputfile, index_col=0, header=0)  # 读取数据
feature = ['x1', 'x3', 'x4', 'x5', 'x6', 'x7', 'x8', 'x13']  # 属性所在列
print('***************分隔*****************')
#data.index = range(1994, 2014)
data_train = 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].values  # 属性数据
y_train = data_train['y'].values  # 标签数据
linearsvr = LinearSVR()  # 调用LinearSVR()函数
linearsvr.fit(x_train,y_train)
x = ((data[feature] - data_mean[feature])/data_std[feature]).values  # 预测，并还原结果。
data['y_pred'] = linearsvr.predict(x) * data_std['y'] + data_mean['y']
outputfile = 'D://CourseAssignment//AI//DataPredict//new_reg_data_GM11_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：\n',data[['y','y_pred']])

fig = data[['y','y_pred']].plot(subplots = True, style=['b-o','r-*'])  # 画出预测结果图
plt.title('3009----predict')
plt.show()

 

 

 

 

 

 

04基于ARIMA的财政预测

import pandas as pd
# 参数初始化
import statsmodels.tsa.arima_model

discfile = r'D://CourseAssignment//AI//DataPredict//data.csv'
forecastnum = 3

# 读取数据，指定日期列为指标，pandas自动将“日期”列识别为Datetime格式
data = pd.read_csv(discfile)
x = 'y'
data = data[[x]]

""""
# 时序图
import matplotlib.pyplot as plt
plt.rcParams['font.sans-serif'] = ['SimHei']  # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False  # 用来正常显示负号
x = [
  '1994','1995','1996','1997','1998','1999','2000'
  ,'2001','2002','2003','2004','2005','2006','2007','2008','2009',
  '2010','2011','2012','2013']
y = data['y']
plt.figure(figsize=(14,9))
plt.plot(x,y,color='green')
plt.show()
"""

# 自相关图
from statsmodels.graphics.tsaplots import plot_acf
plot_acf(data).show()

# 平稳性检测
from statsmodels.tsa.stattools import adfuller as ADF
print('原始序列的ADF检验结果为：', ADF(data['y']))
# 返回值依次为adf、pvalue、usedlag、nobs、critical values、icbest、regresults、resstore

# 差分后的结果
D_data = data.diff().dropna()
D_data.columns = ['收入差分']
D_data.plot()  # 时序图
import matplotlib.pyplot as plt
plt.show()
plot_acf(D_data).show()  # 自相关图
from statsmodels.graphics.tsaplots import plot_pacf
plot_pacf(D_data, lags=8).show()  # 偏自相关图
print('差分序列的ADF检验结果为：', ADF(D_data['收入差分']))  # 平稳性检测

# 白噪声检验
from statsmodels.stats.diagnostic import acorr_ljungbox
print('差分序列的白噪声检验结果为：', acorr_ljungbox(D_data, lags=1))  # 返回统计量和p值

from statsmodels.tsa.arima.model import ARIMA

# 定阶
data['y'] = data['y'].astype('float64')
pmax = int(len(D_data)/10)  # 一般阶数不超过length/10
qmax = int(len(D_data)/10)  # 一般阶数不超过length/10
bic_matrix = []  # BIC矩阵
for p in range(pmax+1):
  tmp = []
  for q in range(qmax+1):
    try:  # 存在部分报错，所以用try来跳过报错。
      tmp.append(ARIMA(data,order= (p,1,q)).fit().bic)
    except:
      tmp.append(None)
  bic_matrix.append(tmp)
bic_matrix = pd.DataFrame(bic_matrix)  # 从中可以找出最小值
p,q = bic_matrix.stack().idxmin()  # 先用stack展平，然后用idxmin找出最小值位置。
print('BIC最小的p值和q值为：%s、%s' %(p,q))
model = ARIMA(data, order=(p,1,q)).fit()  # 建立ARIMA(0, 1, 1)模型
print('模型报告为：\n', model.summary())
print('预测未来3年，其预测结果、标准误差、置信区间如下：\n', model.forecast(3))

#原数据cxcel文件列名规范
data.to_excel('D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_origin.xls')
columns = ['year', 'y']
originfile = 'D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_origin.xls'
origindata = pd.read_excel(originfile, header = None, names = columns)
origindata = origindata.drop(labels=0)
origindata.to_excel('D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_origin.xls')


#提取预测后的数值到excel
model.forecast(3).to_excel('D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_temp.xls')
tempfile = r'D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_temp.xls'
tempdata = pd.read_excel(tempfile)
tempdata.rename(columns={'predicted_mean':'y'}, inplace=True)
tempdata.to_excel('D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_temp.xls')
tempdata.rename(columns={'Unnamed: 0':'year'}, inplace=True)
tempdata.to_excel('D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_temp.xls')


#合并excel
df1 = pd.read_excel('D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_origin.xls')
df2 = pd.read_excel('D://CourseAssignment//AI//DataPredict//ARIMA_data//new_reg_data_ARIMA_temp.xls')
result = pd.concat([df1, df2])
print(result)
result.to_excel('D://CourseAssignment//AI//DataPredict//new_reg_data_ARIMA_result.xls')

# 时序图
plt.rcParams['font.sans-serif'] = ['SimHei']  # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False  # 用来正常显示负号
plt.figure(figsize=(10,4))
x = [
  '1994','1995','1996','1997','1998','1999','2000'
  ,'2001','2002','2003','2004','2005','2006','2007','2008','2009',
  '2010','2011','2012','2013','2014','2015','2016']
y = result['y']
x_pred = ['2014','2015','2016']
y_pred = tempdata['y']
plt.plot(x, y, color = 'green', label = '现有数据', marker='o')
plt.plot(x_pred, y_pred, color = 'red', label = '预测数据', marker='s')
plt.legend()
plt.title('3009---ARIMA预测')
plt.show()