同花顺Supermind量化交易 机器算法运用--预测指数涨跌的算法对比 附源代码

这一节学习机器学习算法的对比，第三章就九大不同的算法对预测指数涨跌情况进行对比。

dt = pd.DataFrame(columns = label)
for date in tradelist:
    stock = get_index_stocks(indexcode,date)
    df = get_price(stock, date, date, '1d', ['quote_rate'], skip_paused = False, fq = 'pre', bar_count = 0, is_panel = 1)['quote_rate'].T.fillna(0)
    label3 = round(df.mean()[0],3)
    label2 = (len(list(df[df[date]>0][date]))-len(list(df[df[date]<0][date])))/len(list(df[date]))
    moneydf = get_money_flow_step(stock,date,date,'1d',['net_flow_rate'],None,is_panel=1)['net_flow_rate'].T.fillna(0)
    label1 = round(moneydf.mean()[0],3)
    dt.loc[date] = [label1,label2,label3]
value = list(get_price(indexcode, startdate, enddate, '1d', ['quote_rate'], skip_paused = False, fq = 'pre', bar_count = 0, is_panel = 1)['quote_rate'])
dt['now up']=value
dt['now label'] = dt['now up'].apply(lambda x:1 if x>0 else -1)
dt['next up']=list(dt['now up'])[1:]+[0]
dt['label']=dt['next up'].apply(lambda x:1 if x>0 else -1)
dt

Out[208]:

	money rate %	net up rate %	mean of updown %	now up	now label	next up	label
20140102	-6.977	-0.44	-0.455	-0.8688	-1	-1.6035	-1
20140103	-14.628	-0.58	-1.197	-1.6035	-1	-1.5062	-1
20140106	-10.977	-0.62	-2.111	-1.5062	-1	-0.1504	-1
20140107	-2.380	-0.32	0.090	-0.1504	-1	0.3097	1
20140108	-4.171	-0.04	-0.175	0.3097	1	-0.7557	-1
20140109	-4.399	-0.62	-0.917	-0.7557	-1	-0.3732	-1
20140110	-8.866	-0.22	-0.622	-0.3732	-1	-0.2439	-1
20140113	-11.486	-0.02	-0.250	-0.2439	-1	0.3126	1
20140114	-7.618	0.32	0.199	0.3126	1	-0.7187	-1
20140115	-12.046	-0.52	-0.620	-0.7187	-1	0.3080	1
20140116	-0.555	0.04	0.262	0.3080	1	-1.1669	-1
20140117	-11.065	-0.64	-1.425	-1.1669	-1	-0.7338	-1
20140120	-10.974	-0.64	-0.878	-0.7338	-1	0.9586	1
20140121	-3.121	0.80	0.927	0.9586	1	2.3827	1
20140122	10.358	1.00	2.710	2.3827	1	-0.9966	-1
20140123	-10.280	-0.72	-0.755	-0.9966	-1	0.2024	1
20140124	0.982	0.48	0.592	0.2024	1	-1.5445	-1
20140127	-12.541	-0.72	-1.411	-1.5445	-1	0.6044	1
20140128	-5.032	0.34	0.366	0.6044	1	0.4495	1
20140129	-7.350	0.08	0.095	0.4495	1	-1.1472	-1
20140130	-13.810	-0.76	-1.215	-1.1472	-1	-0.1355	-1
20140207	-9.088	-0.16	0.115	-0.1355	-1	2.0667	1
20140210	7.470	0.96	2.693	2.0667	1	1.4934	1
20140211	5.662	0.72	1.398	1.4934	1	-0.0864	-1
20140212	-5.996	-0.24	-0.027	-0.0864	-1	0.0664	1
20140213	-4.087	-0.30	-0.516	0.0664	1	0.4073	1
20140214	-7.007	0.36	0.796	0.4073	1	0.2520	1
20140217	-8.357	0.38	0.597	0.2520	1	-1.9221	-1
20140218	-22.680	-0.84	-1.890	-1.9221	-1	1.6476	1
20140219	1.759	0.78	1.418	1.6476	1	-0.8646	-1
...	...	...	...	...	...	...	...
20181211	-5.442	0.30	0.410	0.2894	1	0.3204	1
20181212	-2.328	0.60	0.376	0.3204	1	1.4272	1
20181213	4.618	0.88	1.545	1.4272	1	-1.4046	-1
20181214	-9.521	-0.94	-1.470	-1.4046	-1	0.1058	1
20181217	-5.410	0.46	0.399	0.1058	1	-1.1574	-1
20181218	-9.453	-0.72	-1.188	-1.1574	-1	-1.1810	-1
20181219	-12.110	-0.70	-1.012	-1.1810	-1	-1.4613	-1
20181220	-8.132	-0.64	-1.128	-1.4613	-1	-1.2324	-1
20181221	-7.126	-0.60	-1.204	-1.2324	-1	0.1339	1
20181224	0.890	0.16	0.319	0.1339	1	-0.5153	-1
20181225	-1.902	-0.46	-0.801	-0.5153	-1	-0.6867	-1
20181226	-2.723	-0.44	-0.569	-0.6867	-1	-0.2411	-1
20181227	-4.277	0.02	-0.454	-0.2411	-1	0.7487	1
20181228	1.770	0.48	0.564	0.7487	1	-1.3217	-1
20190102	-5.086	-0.64	-1.368	-1.3217	-1	0.2851	1
20190103	0.290	0.18	0.151	0.2851	1	2.0008	1
20190104	7.130	0.90	2.067	2.0008	1	-0.0141	-1
20190107	-4.997	0.08	0.212	-0.0141	-1	-0.3953	-1
20190108	-2.920	-0.34	-0.415	-0.3953	-1	1.1951	1
20190109	5.368	0.70	1.068	1.1951	1	-0.0372	-1
20190110	-4.994	-0.48	-0.376	-0.0372	-1	0.9712	1
20190111	0.116	0.74	0.777	0.9712	1	-0.9923	-1
20190114	-6.029	-0.62	-0.862	-0.9923	-1	2.0262	1
20190115	6.696	0.90	1.631	2.0262	1	0.1198	1
20190116	-3.404	0.16	-0.150	0.1198	1	-0.4145	-1
20190117	-2.421	-0.38	-0.432	-0.4145	-1	1.9405	1
20190118	4.543	0.86	1.558	1.9405	1	0.6256	1
20190121	-0.123	0.32	0.640	0.6256	1	-1.2849	-1
20190122	-10.820	-0.82	-1.257	-1.2849	-1	-0.1742	-1
20190123	-4.998	-0.02	-0.088	-0.1742	-1	0.0000	-1

1236 rows × 7 columns

KNN

In [256]:

label = ['money rate %','net up rate % ','mean of updown %']
label1 = 'money rate %'
label2 = 'net up rate % '
label3 = 'mean of updown %'

#保留近一年的数据，用于测试，之前数据用于训练
train = dt[:-250]
test = dt[-250:]
X=train[label]
Y=train['label']
X_test=test[label]
Y_test=test['label']

from sklearn.neighbors import KNeighborsClassifier
from sklearn.linear_model import LogisticRegression
from sklearn import tree
from sklearn import svm
from sklearn.naive_bayes import GaussianNB
from sklearn.cluster import KMeans
from sklearn.ensemble import RandomForestClassifier
from sklearn.ensemble import GradientBoostingClassifier
from xgboost import XGBClassifier
dr = pd.DataFrame()
for s in ['tree','KNN','SVM','XGBClass','GBM', 'Random Forest','KMeans','GaussianNB','Logistic']:
    if s == 'tree':
        model = tree.DecisionTreeClassifier() 
    elif s == 'KNN':
        model=KNeighborsClassifier(n_neighbors=30)
    elif s == 'SVM':
        model = svm.SVC() 
    elif s == 'XGBClass':
        model=XGBClassifier()
    elif s == 'GBM':
        model= GradientBoostingClassifier()
    elif s == 'Random Forest':
        model= RandomForestClassifier()
    elif s == 'KMeans':
        model = KMeans()
    elif s == 'GaussianNB':
        model= GaussianNB()
    elif s == 'Logistic':
        model = LogisticRegression()


    model.fit(X, Y)
    print('训练时,预测成功率 {}'.format(round(np.mean(model.predict(X)==Y),2)))
    print('测试时,预测成功率 {}'.format(round(np.mean(model.predict(X_test)==Y_test),2)))

    name = str(s)+' net value'
    #净值
    test['Forecast'] = list(model.predict(X_test))
    test['ref'] = test['next up'].loc[test['Forecast']==1]
    test = test.fillna(0)
    test['ref'] = test['ref'].apply(lambda x:1+x/100)
    from operator import mul
    from functools import reduce
    test['date'] = test.index
    test[name] = test['date'].apply(lambda x:reduce(mul,list(test['ref'])[:list(test['date']).index(x)+1]))
    dr[name] = test[name]
dr

训练时,预测成功率 1.0
测试时,预测成功率 0.53
训练时,预测成功率 0.56
测试时,预测成功率 0.55
训练时,预测成功率 0.61
测试时,预测成功率 0.58
训练时,预测成功率 0.72
测试时,预测成功率 0.5
训练时,预测成功率 0.76
测试时,预测成功率 0.51
训练时,预测成功率 0.97
测试时,预测成功率 0.55
训练时,预测成功率 0.1
测试时,预测成功率 0.09
训练时,预测成功率 0.51
测试时,预测成功率 0.5
训练时,预测成功率 0.53
测试时,预测成功率 0.52

Out[256]:

	tree net value	KNN net value	SVM net value	XGBClass net value	GBM net value	Random Forest net value	KMeans net value	GaussianNB net value	Logistic net value
20180115	1.004937	1.000000	1.004937	1.000000	1.000000	1.000000	1.000000	1.004937	1.004937
20180116	1.008978	1.004021	1.008978	1.004021	1.004021	1.004021	1.000000	1.004937	1.008978
20180117	1.018939	1.004021	1.018939	1.013934	1.013934	1.013934	1.009873	1.004937	1.008978
20180118	1.022635	1.004021	1.018939	1.017611	1.017611	1.013934	1.009873	1.004937	1.012637
20180119	1.026738	1.004021	1.018939	1.021694	1.021694	1.013934	1.009873	1.004937	1.016700
20180122	1.041883	1.018831	1.018939	1.036765	1.036765	1.028890	1.009873	1.019761	1.031697
20180123	1.043139	1.018831	1.018939	1.036765	1.036765	1.028890	1.009873	1.019761	1.031697
20180124	1.035637	1.011504	1.011611	1.029308	1.029308	1.021490	1.009873	1.019761	1.024277
20180125	1.040662	1.011504	1.016520	1.029308	1.029308	1.021490	1.009873	1.024709	1.029247
20180126	1.023578	0.994899	1.016520	1.012411	1.012411	1.021490	1.009873	1.024709	1.029247
20180129	1.009941	0.981644	1.016520	0.998923	0.998923	1.007881	1.009873	1.011057	1.015535
20180130	1.022312	0.981644	1.028971	1.011159	1.011159	1.007881	1.009873	1.023441	1.027974
20180131	1.030313	0.989327	1.037025	1.011159	1.011159	1.015770	1.009873	1.023441	1.036020
20180201	1.033072	0.991977	1.039802	1.013867	1.013867	1.018490	1.012577	1.023441	1.036020
20180202	1.033072	1.001960	1.050266	1.024070	1.013867	1.018490	1.012577	1.033741	1.046446
20180205	1.033072	0.981851	1.029188	1.003517	0.993518	0.998049	0.992255	1.033741	1.025444
20180206	1.003947	0.954169	1.000172	0.975225	0.965508	0.969911	0.964280	1.004597	0.996534
20180207	1.003947	0.954169	1.000172	0.948264	0.938816	0.969911	0.964280	0.976824	0.968984
20180208	0.957660	0.954169	1.000172	0.904544	0.895532	0.969911	0.964280	0.931787	0.924309
20180209	0.956019	0.954169	1.000172	0.904544	0.895532	0.969911	0.964280	0.930190	0.922724
20180212	0.972209	0.970328	1.000172	0.904544	0.910697	0.986336	0.964280	0.945943	0.938351
20180213	0.978708	0.976815	1.006858	0.904544	0.910697	0.992930	0.970727	0.945943	0.938351
20180214	0.978708	0.976815	1.006858	0.904544	0.910697	0.992930	0.970727	0.945943	0.957987
20180222	0.987150	0.976815	1.015543	0.912347	0.918553	1.001495	0.979100	0.945943	0.957987
20180223	0.992697	0.976815	1.021249	0.912347	0.918553	1.007122	0.984602	0.945943	0.957987
20180226	0.992697	0.976815	1.021249	0.912347	0.918553	1.007122	0.984602	0.945943	0.957987
20180227	0.992697	0.976815	1.004353	0.897252	0.903356	0.990460	0.984602	0.930292	0.942137
20180228	0.997053	0.981101	1.008760	0.901189	0.907320	0.994806	0.984602	0.934374	0.946271
20180301	0.997053	0.981101	1.008760	0.901189	0.907320	0.985476	0.984602	0.925612	0.937397
20180302	0.996986	0.981101	1.008760	0.901129	0.907259	0.985476	0.984602	0.925550	0.937334
...	...	...	...	...	...	...	...	...	...
20181211	0.965227	1.098993	1.191051	0.903900	0.912769	0.986589	0.987298	0.882477	0.913479
20181212	0.965227	1.114678	1.208050	0.903900	0.925796	1.000669	0.987298	0.882477	0.913479
20181213	0.965227	1.114678	1.208050	0.891203	0.925796	1.000669	0.973431	0.882477	0.913479
20181214	0.965227	1.115857	1.209328	0.892146	0.926776	1.001728	0.973431	0.883411	0.914446
20181217	0.965227	1.115857	1.209328	0.881821	0.916049	0.990134	0.973431	0.883411	0.914446
20181218	0.965227	1.102679	1.195046	0.871406	0.905231	0.990134	0.973431	0.872978	0.903646
20181219	0.951122	1.086565	1.177583	0.871406	0.892003	0.990134	0.973431	0.860221	0.890441
20181220	0.951122	1.086565	1.163070	0.871406	0.892003	0.990134	0.973431	0.849620	0.879468
20181221	0.952395	1.086565	1.163070	0.871406	0.892003	0.990134	0.973431	0.850757	0.880645
20181224	0.947488	1.086565	1.163070	0.866916	0.887406	0.990134	0.973431	0.850757	0.876107
20181225	0.940981	1.086565	1.155083	0.866916	0.881312	0.990134	0.973431	0.844915	0.870091
20181226	0.938713	1.083946	1.152299	0.866916	0.881312	0.987747	0.973431	0.842878	0.867993
20181227	0.938713	1.083946	1.152299	0.866916	0.881312	0.987747	0.973431	0.849189	0.874492
20181228	0.938713	1.069619	1.137069	0.866916	0.881312	0.987747	0.960565	0.849189	0.874492
20190102	0.938713	1.072669	1.140310	0.866916	0.881312	0.987747	0.960565	0.851610	0.876985
20190103	0.957494	1.072669	1.140310	0.884261	0.898946	1.007510	0.960565	0.851610	0.894532
20190104	0.957359	1.072517	1.140150	0.884137	0.898819	1.007368	0.960565	0.851610	0.894532
20190107	0.957359	1.072517	1.140150	0.884137	0.898819	1.007368	0.960565	0.848243	0.890996
20190108	0.968801	1.085335	1.153776	0.894703	0.909561	1.019407	0.960565	0.858381	0.901644
20190109	0.968801	1.085335	1.153776	0.894703	0.909561	1.019407	0.960208	0.858381	0.901644
20190110	0.978210	1.095876	1.153776	0.903392	0.909561	1.019407	0.960208	0.866717	0.910401
20190111	0.978210	1.095876	1.153776	0.903392	0.909561	1.009291	0.960208	0.866717	0.910401
20190114	0.998030	1.118081	1.153776	0.921697	0.927990	1.029741	0.960208	0.884279	0.928847
20190115	0.999226	1.119420	1.155158	0.922801	0.929102	1.029741	0.960208	0.884279	0.928847
20190116	0.995084	1.114780	1.150370	0.918976	0.925251	1.029741	0.960208	0.880613	0.924997
20190117	1.014394	1.136412	1.172693	0.936809	0.943205	1.049723	0.960208	0.897702	0.942947
20190118	1.014394	1.136412	1.172693	0.942669	0.943205	1.049723	0.966215	0.897702	0.942947
20190121	1.001360	1.136412	1.172693	0.930557	0.931086	1.036236	0.966215	0.897702	0.930831
20190122	0.999615	1.134433	1.170650	0.930557	0.929464	1.034430	0.966215	0.896138	0.929209
20190123	0.999615	1.134433	1.170650	0.930557	0.929464	1.034430	0.966215	0.896138	0.929209

250 rows × 9 columns

In [257]:

fig = plt.figure()
axes = fig.add_axes([0.1, 0.1, 1, 1.382]) #插入面板
color = ['tomato','green','darkorchid','lightskyblue','y','gold','deeppink','lightgoldenrodyellow','red']
t = list(dr.columns)
for s in t:
    g = t.index(s)
    x1_list=list(dr[s])
    y=np.array(x1_list)
    x=np.array(range(0,len(x1_list)))
    axes.plot(x, y , color = color[g])

    axes.set_xlabel('Time',fontsize=15)
    axes.set_ylabel('net value',fontsize=15)
    axes.set_title('AI return ',fontsize=20)
    axes.legend(t)
    #设置X轴
    mtradelist = list(test['date'])
    numlist=[]
    for s in list(range(0,len(mtradelist),60)):
        numlist.append(mtradelist[s])
    axes.set_xticks(list(range(0,len(mtradelist),60)))
    axes.set_xticklabels(numlist, fontsize=10)

查看以上策略详细请 到 supermind量化交易官网查看：机器算法运用--预测指数涨跌的算法对比 附源代码