利用ARIMA模型來進行時間序列的預測，數據集是雅虎財經的離線股票數據

import pandas as pd
import pandas_datareader
import datetime
import matplotlib.pyplot as plt
import seaborn as sns
from matplotlib.pylab import style
from statsmodels.tsa.arima_model import ARIMA
from statsmodels.graphics.tsaplots import plot_acf, plot_pacf

style.use(
 
‘ggplot‘)
plt.rcParams[‘font.sans-serif‘] = [‘SimHei‘]#用來正常顯示中文標簽
plt.rcParams[‘axes.unicode_minus‘] = False#用來正常顯示負號

stock = pd.read_csv(‘D:/T10yr.csv‘, index_col=0, parse_dates=[0])
# print(stock.head())

stock_week = stock[‘Close‘].resample(‘W-MON‘).mean()
stock_train = stock_week[‘2000‘ : ‘2012‘]
# print(stock_train)
 


stock_train.plot(figsize=(12, 8))
plt.legend(bbox_to_anchor=(1.25, 0.5))
plt.title(‘Stock Close‘)
sns.despine()# 刪除上方和右方的邊框
# plt.show()

#數據差異過大，進行一階差分
stock_diff = stock_train.diff()
stock_diff = stock_diff.dropna()
print(stock_diff)
plt.figure()#新生成一塊畫布
plt.plot(stock_diff)
plt.title(‘一階差分‘)
# plt.show()
 


#畫圖ACF確定幾階
acf = plot_acf(stock_diff, lags=20)
plt.title(‘ACF‘)
# plt.show()
#PACF
pacf = plot_pacf(stock_diff, lags=20)
plt.title(‘PACF‘)
# plt.show()

#ARIMA model
model = ARIMA(stock_train, order=(1, 1, 1), freq=‘W-MON‘)
result = model.fit()
# print(result.summary())

pred = result.predict(‘20110103‘, ‘20160701‘, dynamic=True, typ=‘levels‘)
print(pred)

plt.figure(figsize=(6, 6))
plt.plot(pred)
plt.plot(stock_week)
plt.show()

股票預測