arima时间序列预测

代码一：

from sklearn.metrics import r2_score 
import pandas as pd
def evaluate_arima_model(X, arima_order):
	# prepare training dataset
	train_size = int(len(X) * 0.66)
	train, test = X[0:train_size], X[train_size:]
	history = [x for x in train]
	# make predictions
	predictions = list()
	for t in range(len(test)):
		model = ARIMA(history, order=arima_order)
		model_fit = model.fit(disp=0)
		yhat = model_fit.forecast()[0]
		predictions.append(yhat)
		history.append(test[t])
	# calculate out of sample error
	error = r2_score(test, predictions)
	return error

# evaluate combinations of p, d and q values for an ARIMA model
def evaluate_models(dataset, p_values, d_values, q_values):
	dataset = dataset.astype('float32')
	best_score, best_cfg = float("inf"), None
	for p in p_values:
		for d in d_values:
			for q in q_values:
				order = (p,d,q)
				try:
					r2 = evaluate_arima_model(dataset, order)
					if r2 > best_score:
						best_score, best_cfg = r2, order
					print('ARIMA%s MSE=%.3f' % (order,r2))
				except:
					continue
	print('Best ARIMA%s R2=%.3f' % (best_cfg, best_score))

# load dataset

sheet= pd.read_excel('F:\\123123.xlsx',sheet_name= [0,1,2,3],header=0, index_col=0)[0]
series=sheet["XXX"]
# evaluate parameters
p_values = [0, 1, 2, 4, 6, 8, 10,12,14,16]
d_values = range(0, 5)
q_values = range(0, 5)
warnings.filterwarnings("ignore")
evaluate_models(series.values, p_values, d_values, q_values)

 

代码二：

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from sklearn.preprocessing import LabelEncoder
from sklearn.preprocessing import MinMaxScaler
from sklearn.metrics import r2_score
from pandas import DataFrame
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import LSTM
from pandas import concat
from numpy import concatenate
from sklearn.metrics import mean_squared_error
from math import sqrt
from matplotlib import pyplot
sheet = pd.read_excel('F:\\123123.xlsx',sheet_name= [0,1,2,3],index_col=u'日期')

data=sheet[0][["XXX]].copy()
#绘制时序图
#plt.plot(date,recharge)
plt.rcParams['font.sans-serif']=['SimHei']
plt.rcParams['axes.unicode_minus']=False
data.plot()
plt.show()

##绘制对数时序图
log_data=np.log(data)
log_data.plot()
plt.show

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


from statsmodels.tsa.stattools import adfuller as ADF
print('原始序列的ADF检测结果为：',ADF(data[u'XXX']))

##差分
D_data=data.diff().dropna()
D_data.columns=[u'XXX']
D_data.plot()
plt.show()
plot_acf(D_data).show()
from statsmodels.graphics.tsaplots import plot_pacf
plot_pacf(D_data).show()
print('差分序列的ADF检测结果为：',ADF(data[u'XXX']))

from statsmodels.stats.diagnostic import acorr_ljungbox
print('差分序列的白噪声检测结果为：',acorr_ljungbox(D_data,lags=1))

from statsmodels.tsa.arima_model import ARIMA
print("开始建立时间序列")

model=ARIMA(data,(22,0,1)).fit()##这里的参数需要根据前面的adf检测后进行选择的参数
#model.summary2()
pr=model.predict()##predict这个不填参数的意思是将所有目前存在的数据都进行预测，我们可以拿来和真
                      #实数据进行对比
#pr2=model.forecast(5)##forecast这个方法可以直接预测后多少天的数据，这里是预测后5天的数据
#print(pr)
r2=r2_score(pr, data)
print('R2: %.3f' % r2)

print(model.predict('2019-8-22','2019-8-28'))##predict可以直接填写日期进行预测，
                                        ##这里是预测8-22至8-28的数据