1. 前言

数据获取，预处理，定阶什么的参考前面几篇文章： 
金融时间序列分析：3. First Demo By Python
金融时间序列分析：5. AR模型实例（Python）

MA模型的讲解参考前文： 
金融时间序列分析：7. MA滑动平均模型

2. 建模与分析预测

这个和以前那个AR模型基本一样，我也不多说了。

    print '---------------------------MA TEST------------------------------'
    adf_res = adf_test(ts_diff['2014-01-01':'2015-12-31'])
    print 'ADF test results (ema diff):\n', adf_res
    ma_predict = draw_ma(ts_diff['2014-01-01':'2015-12-31'], int(adf_res['Lags Used']))
    rtn_test(ma_predict, "ma predict")

def draw_ma(ts, w):
    ma = ARMA(ts, order=(0, w)).fit(disp = -1)
    ts_predict_ma = ma.predict()

    ar = ARMA(ts, order=(w,0)).fit(disp=-1)
    ts_predict_ar = ar.predict()    

    plt.clf()
    plt.plot(ts_predict_ar, label="AR")
    plt.plot(ts_predict_ma, label="MA")
    #plt.plot(ts, label = "ORG")
    plt.legend(loc="best")
    plt.title("MA Test %s" % w)
    plt.savefig("./PDF/test_ma_"+ str(w) +".pdf", format='pdf')

    return ts_predict_ma

结果如下： 

3. 预测数据分析

原始数据 
abs mean : 0.0118623137237 
std var : 0.0172625800497 
Minimum value is: -0.0925615434824 
Maximum value is: 0.0903425141497 
Arithmetic mean is: 0.000195754536494 
Unbiased variance is: 0.000297996669973 
Biased skewness is: -0.479992664412 
Biased kurtosis is: 3.92636403052

AR预测数据 
abs mean : 0.00363097137215 
std var : 0.00522169036148 
Minimum value is: -0.0273872869419 
Maximum value is: 0.0262852580925 
Arithmetic mean is: 0.0010682950804 
Unbiased variance is: 2.72660502312e-05 
Biased skewness is: -0.391757274581 
Biased kurtosis is: 4.91810357904

MA预测数据 
abs mean : 0.00314592965807 
std var : 0.00438791465633 
Minimum value is: -0.0223893454803 
Maximum value is: 0.0186398706818 
Arithmetic mean is: 0.00106361911149 
Unbiased variance is: 1.92537950312e-05 
Biased skewness is: -0.59323473066 
Biased kurtosis is: 3.99711788068