时间序列（五）股票分析

首先导入相关模块

import pandas as pdimport pandas_datareaderimport datetimeimport matplotlib.pylab as pltimport seaborn as snsfrom matplotlib.pylab import stylefrom statsmodels.tsa.arima_model import ARIMAfrom statsmodels.graphics.tsaplots import plot_acf, plot_pacf

设置样式

style.use('ggplot')    plt.rcParams['font.sans-serif'] = ['SimHei'] plt.rcParams['axes.unicode_minus'] = False  

读入数据

stockFile = 'data/T10yr.csv'stock = pd.read_csv(stockFile, index_col=0, parse_dates=[0])stock.head(10)

重采样

stock_week = stock['Close'].resample('W-MON').mean()stock_train = stock_week['2000':'2015']

画图

stock_train.plot(figsize=(12,8))plt.legend(bbox_to_anchor=(1.25, 0.5))plt.title("Stock Close")sns.despine()plt.show()

发现波动太大，于是做一阶差分

stock_diff = stock_train.diff()stock_diff = stock_diff.dropna()plt.figure()plt.plot(stock_diff)plt.title('一阶差分')plt.show()

接下来通过acf,pacf确定模型的p值以及q值

fig=plt.figure()ax1=fig.add_subplot(211)ax2=fig.add_subplot(212)acf = plot_acf(stock_diff, lags=20,ax=ax1,title="ACF")pacf = plot_pacf(stock_diff, lags=20,ax=ax2,title="PACF")plt.show()

可以看出p,q分别取1就可以了

训练

model = ARIMA(stock_train, order=(1, 1, 1),freq='W-MON')result = model.fit()

预测

pred = result.predict('20140609', '20160701',dynamic=True, typ='levels')print (pred)

预测结果
2014-06-09 2.463559
2014-06-16 2.455539
2014-06-23 2.449569
2014-06-30 2.444183
2014-07-07 2.438962
2014-07-14 2.433788
2014-07-21 2.428627
2014-07-28 2.423470
2014-08-04 2.418315
2014-08-11 2.413159
2014-08-18 2.408004
2014-08-25 2.402849
2014-09-01 2.397693
2014-09-08 2.392538
2014-09-15 2.387383
2014-09-22 2.382227
2014-09-29 2.377072
2014-10-06 2.371917
2014-10-13 2.366761
2014-10-20 2.361606
2014-10-27 2.356451
2014-11-03 2.351296
2014-11-10 2.346140
2014-11-17 2.340985
2014-11-24 2.335830
2014-12-01 2.330674
2014-12-08 2.325519
2014-12-15 2.320364
2014-12-22 2.315208
2014-12-29 2.310053
…
2015-12-07 2.057443
2015-12-14 2.052288
2015-12-21 2.047132
2015-12-28 2.041977
2016-01-04 2.036822
2016-01-11 2.031666
2016-01-18 2.026511
2016-01-25 2.021356
2016-02-01 2.016200
2016-02-08 2.011045
2016-02-15 2.005890
2016-02-22 2.000735
2016-02-29 1.995579
2016-03-07 1.990424
2016-03-14 1.985269
2016-03-21 1.980113
2016-03-28 1.974958
2016-04-04 1.969803
2016-04-11 1.964647
2016-04-18 1.959492
2016-04-25 1.954337
2016-05-02 1.949181
2016-05-09 1.944026
2016-05-16 1.938871
2016-05-23 1.933716
2016-05-30 1.928560
2016-06-06 1.923405
2016-06-13 1.918250
2016-06-20 1.913094
2016-06-27 1.907939
Freq: W-MON, Length: 108, dtype: float64

图像展示结果

plt.figure(figsize=(6, 6))plt.xticks(rotation=45)plt.plot(pred)plt.plot(stock_train)plt.show()