├── .project
├── .pydevproject
├── .settings
    └── org.eclipse.core.resources.prefs
├── README.md
├── classification
    ├── __init__.py
    ├── calc_profit.py
    ├── featureSelection.py
    └── predictStock.py
├── clawer
    ├── USAStock_clawer.py
    ├── __init__.py
    └── code_name.py
├── data_download.py
├── getData
    ├── __init__.py
    └── getTrainData.py
└── 美股涨跌预测系统的探究-141499罗斌.docx


/.project:
--------------------------------------------------------------------------------
 1 | <?xml version="1.0" encoding="UTF-8"?>
 2 | <projectDescription>
 3 | 	<name>MLHomework34</name>
 4 | 	<comment></comment>
 5 | 	<projects>
 6 | 	</projects>
 7 | 	<buildSpec>
 8 | 		<buildCommand>
 9 | 			<name>org.python.pydev.PyDevBuilder</name>
10 | 			<arguments>
11 | 			</arguments>
12 | 		</buildCommand>
13 | 	</buildSpec>
14 | 	<natures>
15 | 		<nature>org.python.pydev.pythonNature</nature>
16 | 	</natures>
17 | </projectDescription>
18 | 


--------------------------------------------------------------------------------
/.pydevproject:
--------------------------------------------------------------------------------
1 | <?xml version="1.0" encoding="UTF-8" standalone="no"?>
2 | <?eclipse-pydev version="1.0"?><pydev_project>
3 | <pydev_pathproperty name="org.python.pydev.PROJECT_SOURCE_PATH">
4 | <path>/${PROJECT_DIR_NAME}</path>
5 | </pydev_pathproperty>
6 | <pydev_property name="org.python.pydev.PYTHON_PROJECT_VERSION">python 3.0</pydev_property>
7 | <pydev_property name="org.python.pydev.PYTHON_PROJECT_INTERPRETER">C:\Python34\python.exe</pydev_property>
8 | </pydev_project>
9 | 


--------------------------------------------------------------------------------
/.settings/org.eclipse.core.resources.prefs:
--------------------------------------------------------------------------------
1 | eclipse.preferences.version=1
2 | encoding//clawer/USAStock_clawer.py=utf-8
3 | encoding/<project>=UTF-8
4 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # stockPrediction
2 | 本项目对美股股票的涨跌进行了研究，从问题出发并提出猜想，然后定义了机器学习的实验任务。通过多次实验得到实验数据，最终证明了所提出的猜想：中国股市确实和美国股市存在着一定的联系，并且通过这些隐含的联系可以预测某些美国股票的涨跌。
3 | 


--------------------------------------------------------------------------------
/classification/__init__.py:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/jxls080511/stockPrediction/7ccb30b9f2b3537347e00b613ec0dbbaa97c6bc7/classification/__init__.py


--------------------------------------------------------------------------------
/classification/calc_profit.py:
--------------------------------------------------------------------------------
  1 | import os
  2 | import pickle
  3 | 
  4 | def calc():
  5 |     log_out=open('I:\\MLHomework\\Data\\2\\clf8.log','w',encoding='utf-8')#log文件
  6 |     result_out=open('I:\\MLHomework\\Data\\2\\result8.log','w',encoding='utf-8')#log文件
  7 |     clf_dir='I:\\MLHomework\\Data\\2\\clf8'
  8 |     #先读入5-8月中国股市的数据
  9 |     chinese_stock={}
 10 |     with open('I:\\MLHomework\\Data\\4\\testData4m\\ASX.csv.data','r',encoding='utf-8') as f:
 11 |         for line in f:
 12 |             array=line.split('\t')
 13 |             date=array[0]
 14 |             v=int(array[len(array)-1])#最后的值不需要了
 15 |             l=list(0 for i in range(len(array)-2))
 16 |             for i in range(1,len(array)-2):
 17 |                 l[i]=int(array[i])
 18 |             chinese_stock[date]=l
 19 |     
 20 |     us_stock_record={}
 21 |     for filename in os.listdir(clf_dir):
 22 |         us_stock_record[filename]={}
 23 |         with open('I:\\MLHomework\\stockData_american\\'+filename+'.csv','r',encoding='utf-8') as f:
 24 |             firstline=True#第一行不是数据
 25 |             for line in f:
 26 |                 if(firstline):
 27 |                     firstline=False
 28 |                     continue
 29 |                 else:
 30 |                     array=line.strip().split(',')
 31 |                     date=array[0]
 32 |                     openValue=float(array[1])
 33 |                     closeValue=float(array[4])
 34 |                     l=list(0 for i in range(2))
 35 |                     l[0]=openValue
 36 |                     l[1]=closeValue
 37 |                     us_stock_record[filename][date]=l
 38 |     
 39 |     money=1000000.0
 40 |     #for循环，每一天
 41 |     for year in range(2015,2016):
 42 |         for month in range(5,9):
 43 |             for day in range(1,32):
 44 |                 date=str(year)+'-'
 45 |                 if(month<10):
 46 |                     date+='0'+str(month)+'-'
 47 |                 else:
 48 |                     date+=str(month)+'-'
 49 |                     
 50 |                 if(day<10):
 51 |                     date+='0'+str(day)
 52 |                 else:
 53 |                     date+=str(day)  
 54 |                 #此时的date是一个日期string
 55 |                 
 56 |                 if(date in chinese_stock.keys()):
 57 |                     print(date+'\t'+str(money))
 58 |                     result_out.write(str(money)+'\n')
 59 |                     log_out.write(date+'\t'+str(money)+'\n')
 60 |                     #得到当天的中国股市情况
 61 |                     l=chinese_stock[date]
 62 |                     #对于高准确率的股票进行预测，并记下当天的开盘价收盘价
 63 |                     us_stock={}#记录美股的预测情况
 64 |                     buy_num=0#今天需要买入的股票数量
 65 |                     for stock_name in os.listdir(clf_dir):
 66 |                         record=list(0 for i in range(3))
 67 |                         #读入分类器
 68 |                         with open(clf_dir+'\\'+stock_name,'br') as f:
 69 |                             clf=pickle.load(f)
 70 |                             result=clf.predict(l)
 71 |                             if(1==result):
 72 |                                 buy_num+=1
 73 |                             try:
 74 |                                 record[0]=int(result[0])#第一列存预测结果
 75 |                                 record[1]=us_stock_record[stock_name][date][0]#开盘价
 76 |                                 record[2]=us_stock_record[stock_name][date][1]#收盘价
 77 |                             except:
 78 |                                 #print('error')
 79 |                                 #buy_num=0#如果出现异常，就不交易了，日期报错
 80 |                                 record[0]=0;
 81 |                                 if(1==result):
 82 |                                     buy_num-=1
 83 |                         us_stock[stock_name]=record
 84 |                     
 85 |                     #计算今天的收益
 86 |                     if(buy_num!=0):
 87 |                         div_money=(money+0.0)/buy_num
 88 |                         today_sum_money=0.0
 89 |                         for stock_name in us_stock.keys():
 90 |                             if(us_stock[stock_name][0]==1):
 91 |                                 p=(us_stock[stock_name][2]+0.0)/us_stock[stock_name][1]#收益率
 92 |                                 earn=div_money*(p-1.0)
 93 |                                 today_sum_money+=div_money*p
 94 |                                 log_out.write(stock_name+'\t'+str(us_stock[stock_name][1])+'\t'+str(us_stock[stock_name][2])+'\t'+str(earn)+'\n')
 95 |                         #以收盘价卖出所有股票
 96 |                         money=today_sum_money
 97 |     log_out.close()
 98 |     result_out.close()
 99 |     return money                
100 |     
101 |         
102 |     
103 |     #返回最后剩下的钱
104 |     
105 | 
106 | if __name__ == '__main__':
107 |     print(calc())


--------------------------------------------------------------------------------
/classification/featureSelection.py:
--------------------------------------------------------------------------------
 1 | from sklearn import datasets
 2 | from sklearn.feature_selection import RFE
 3 | from sklearn.linear_model import LogisticRegression
 4 | from sklearn.ensemble import RandomForestClassifier
 5 | from sklearn import svm
 6 | 
 7 | def test_data(clf,testFile,remove_list):
 8 |     sum=0
 9 |     right_sum=0
10 |     with open(testFile,'r',encoding='utf-8') as f:
11 |         for line in f:
12 |             sum+=1
13 |             array=line.split('\t')
14 |             v=int(array[len(array)-1])
15 |             l=[]
16 |             for i in range(1,len(array)-1):
17 |                 if(i-1 not in remove_list):
18 |                     l.append(int(array[i]))
19 |             
20 |             #print(len(l))   
21 |             result=clf.predict(l)
22 |             if(result==v):
23 |                 right_sum+=1
24 |             #print(str(v)+'\t'+str(result))
25 |     print('precision='+str(right_sum/(sum+0.0)))
26 |     print(str(right_sum)+'/'+str(sum))
27 |     return right_sum/(sum+0.0)
28 | 
29 | if __name__ == '__main__':
30 |     #读文件
31 |     data=[]
32 |     target=[]
33 |     with open('I:\\MLHomework\\Data\\2\\trainData1y\\NOBGY.csv.data','r',encoding='utf-8') as f:
34 |         for line in f:
35 |             array=line.split('\t')
36 |             #print(len(array))
37 |             l=list(0 for i in range(len(array)-2))
38 |             for i in range(1,len(array)-1):
39 |                 l[i-1]=int(array[i])
40 |             data.append(l)
41 |             #print(len(l))  
42 |             target.append(int(array[len(array)-1]))
43 |     
44 |             
45 |     # create a base classifier used to evaluate a subset of attributes
46 |     #model = RandomForestClassifier()
47 |     remove_list=[]
48 |     
49 |     print('RandomForestClassifier')
50 |     clf=RandomForestClassifier()
51 |     clf.fit(data, target)
52 |     test_data(clf,'I:\\MLHomework\\Data\\2\\testData8m\\NOBGY.csv.data',remove_list)
53 |     print('SVM')
54 |     clf11=svm.SVC(kernel='rbf')
55 |     clf11.fit(data, target)
56 |     test_data(clf11,'I:\\MLHomework\\Data\\2\\testData8m\\NOBGY.csv.data',remove_list)
57 |     
58 |     print('1')
59 |     for i in range(0,len(clf.feature_importances_)):
60 |         if(clf.feature_importances_[i]==0.0):
61 |             remove_list.append(i)
62 |     print(len(remove_list))
63 |     #读文件
64 |     data2=[]
65 |     target2=[]
66 |     with open('I:\\MLHomework\\Data\\2\\trainData1y\\NOBGY.csv.data','r',encoding='utf-8') as f:
67 |         for line in f:
68 |             array=line.split('\t')
69 |             l=[]
70 |             for i in range(1,len(array)-2):
71 |                 if(i-1 not in remove_list):
72 |                     l.append(int(array[i]))
73 |             data2.append(l)
74 |             target2.append(int(array[len(array)-1]))
75 |             
76 |     print('RandomForestClassifier')
77 |     clf2=RandomForestClassifier()
78 |     clf2.fit(data2, target2)
79 |     test_data(clf2,'I:\\MLHomework\\Data\\2\\testData8m\\NOBGY.csv.data',remove_list)
80 |     print('SVM')
81 |     clf22=svm.SVC(kernel='rbf')
82 |     clf22.fit(data2, target2)
83 |     test_data(clf22,'I:\\MLHomework\\Data\\2\\testData8m\\NOBGY.csv.data',remove_list)
84 | #     for importance in clf.feature_importances_:
85 | #         print(importance)
86 |         
87 | 


--------------------------------------------------------------------------------
/classification/predictStock.py:
--------------------------------------------------------------------------------
  1 | from sklearn.naive_bayes import GaussianNB
  2 | from sklearn import tree
  3 | from sklearn import svm
  4 | from sklearn.ensemble import RandomForestClassifier
  5 | from sklearn.naive_bayes import GaussianNB
  6 | 
  7 | import os
  8 | import pickle
  9 | 
 10 | def train_data(filePath):
 11 |     data=[]
 12 |     target=[]
 13 |     with open(filePath,'r',encoding='utf-8') as f:
 14 |         for line in f:
 15 |             array=line.split('\t')
 16 |             l=list(0 for i in range(len(array)-2))
 17 |             for i in range(1,len(array)-2):
 18 |                 l[i]=int(array[i])
 19 |             data.append(l)
 20 |             target.append(int(array[len(array)-1]))
 21 |     
 22 |     #clf=RandomForestClassifier()
 23 |     try:
 24 |         clf=svm.SVC(kernel='rbf')
 25 |         #clf=RandomForestClassifier()
 26 |         #clf=GaussianNB()
 27 |         #clf=tree.DecisionTreeClassifier()
 28 |         
 29 |         clf.fit(data, target)
 30 |         return clf
 31 |     except:
 32 |         return None
 33 |     
 34 | 
 35 | 
 36 | 
 37 | def test_data(clf,testFile):
 38 |     sum=0
 39 |     right_sum=0
 40 |     with open(testFile,'r',encoding='utf-8') as f:
 41 |         for line in f:
 42 |             sum+=1
 43 |             array=line.split('\t')
 44 |             v=int(array[len(array)-1])
 45 |             l=list(0 for i in range(len(array)-2))
 46 |             for i in range(1,len(array)-2):
 47 |                 l[i]=int(array[i])
 48 |             result=clf.predict(l)
 49 |             if(result==v):
 50 |                 right_sum+=1
 51 |             print(str(v)+'\t'+str(result))
 52 |     print('precision='+str(right_sum/(sum+0.0)))
 53 |     print(str(right_sum)+'/'+str(sum))
 54 |     return right_sum/(sum+0.0)
 55 | 
 56 | def perdict_all():
 57 |     trainDir='I:\\MLHomework\\\Data\\4\\trainData4m'
 58 |     testDir='I:\\MLHomework\\\Data\\4\\testData4m'
 59 |     clfDir='I:\\MLHomework\\\Data\\2\\clf8'#分类器保存位置
 60 |     
 61 |     out=open('I:\\MLHomework\\\Data\\2\\result_svm3.txt','w',encoding='utf-8')
 62 |     #数据统计
 63 |     sum=0
 64 |     pn_list=[0,0,0,0,0,0,0,0,0,0]#分别表示准确率50-60的，60-70的，70-80的，80-90的，90-100的数量
 65 |     #训练数据文件夹
 66 |     for filename in os.listdir(trainDir):
 67 |         #测试数据文件夹
 68 |         if(filename in os.listdir(testDir)):
 69 |             print(filename[0:len(filename)-9])
 70 |             clf=train_data(trainDir+'\\'+filename)
 71 |             
 72 |             if(clf!=None):
 73 |                 p=test_data(clf,testDir+'\\'+filename)
 74 |                 sum+=1
 75 |                 #把这个股票代码保存起来
 76 |                 out.write(filename[0:len(filename)-9]+'\t'+str(p)+'\n')
 77 | 
 78 |                 if(p>=0.8):
 79 |                     #把分类器保存起来
 80 |                     with open(clfDir+'\\'+filename[0:len(filename)-9],'bw') as f:
 81 |                         pickle.dump(clf,f)
 82 | #                     if(os.path.exists(trainDir+'\\'+filename)):
 83 | #                         os.remove(trainDir+'\\'+filename)
 84 | #                     if(os.path.exists(testDir+'\\'+filename)):
 85 | #                         os.remove(testDir+'\\'+filename)
 86 |                 if(p==1):#防止在1的时候数组出界
 87 |                     p-=0.01
 88 | 
 89 |                 pn_list[int((p*100)/10)]=pn_list[int((p*100)/10)]+1
 90 |                 
 91 |                 for i in range(10):
 92 |                     print(str(i*10)+'~'+str(10+i*10)+'='+str(pn_list[i])+'\t\t\t占'+str(pn_list[i]/(sum+0.0)))
 93 |     out.close()
 94 | 
 95 | def perdict_193Stock(clfDir,testDataPath,outFilePath):
 96 |     l=list(0 for i in range(2892))#一共中国股票是2892维/个
 97 |     with open(testDataPath,'r',encoding='utf-8') as f:
 98 |         for line in f:
 99 |             array=line.split('\t')
100 |             for i in range(2892):
101 |                 l[i]=array[i]
102 |             
103 |             
104 |     out=open(outFilePath,'w',encoding='utf-8')
105 |     for filename in os.listdir(clfDir):
106 |         print(filename)
107 |         with open(clfDir+'\\'+filename,'br') as f:
108 |             clf=pickle.load(f)
109 |             result=clf.predict(l)
110 |             out.write(filename+'\t'+str(result[0])+'\n')
111 |     out.close()
112 |     
113 | 
114 | if __name__ == '__main__':
115 |     #perdict_193Stock('I:\\MLHomework\\Data\\2\\clf','I:\\MLHomework\\1dayData.txt','I:\\MLHomework\\1dayDataResult.txt')
116 |     perdict_all()


--------------------------------------------------------------------------------
/clawer/USAStock_clawer.py:
--------------------------------------------------------------------------------
 1 | # -*- coding: utf-8 -*- #
 2 | import re
 3 | import requests
 4 | import urllib
 5 | import time
 6 | import os
 7 | from bs4 import BeautifulSoup
 8 | import re
 9 | 
10 | def get_encoding(text):
11 |     try:
12 |         return re.search('charset=\"?([^\"]+)"',text).group(1)
13 |     except:
14 |         return None
15 | 
16 | 
17 | def clawer_page(url):
18 |     try:
19 |         response = requests.get(url)
20 |         text=response.text
21 |         encoding=get_encoding(text)
22 |         if(encoding is not None):
23 |             response.encoding=encoding
24 |         text=response.text
25 |         return text
26 |     except:
27 |         print('error')
28 |         return None
29 | 
30 |     
31 | 
32 | 
33 | if __name__ == '__main__':
34 |     for i in range(8):
35 |         url='http://data.tsci.com.cn/US/USCODE.aspx?M=3&First=All&Sid=&uhu=df&P='+str(i)
36 |         page=clawer_page(url)
37 |         out=open('D:\\programing\\Python\\MLHomework\\美股数据\\美国证券交易所'+str(i),'w',encoding='utf-8')
38 |         out.write(page)
39 |         out.close()
40 |         print(i)
41 |         


--------------------------------------------------------------------------------
/clawer/__init__.py:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/jxls080511/stockPrediction/7ccb30b9f2b3537347e00b613ec0dbbaa97c6bc7/clawer/__init__.py


--------------------------------------------------------------------------------
/clawer/code_name.py:
--------------------------------------------------------------------------------
 1 | import re
 2 | 
 3 | def china():
 4 |     out=open('D:\\programing\\Python\\MLHomework\\american_codeName.txt','w',encoding='utf-8')
 5 |     with open('D:\\programing\\Python\\MLHomework\\american_codeNameaaa.txt','r',encoding='utf-8') as f:
 6 |         for line in f:
 7 |             array=line.strip().split('\t')
 8 |             match=re.search('([^(]+)\\(([^\\)]+)\\)',array[0])
 9 |             name=match.group(1)
10 |             code=match.group(2)
11 |             market=array[2]
12 |             out.write(name+'\t'+code+'\t'+market+'\n')
13 |     out.close()
14 |                 
15 | 
16 | def american():
17 |     out=open('D:\\programing\\Python\\MLHomework\\american_codeName2.txt','w',encoding='utf-8')
18 |     with open('D:\\programing\\Python\\MLHomework\\美股抽取.txt','r',encoding='utf-8') as f:
19 |         for line in f:
20 |             array=line.strip().split('\t')
21 |             name=''
22 |             code=''
23 |             market=''
24 |             if('.NASDQ' in array[0]):
25 |                 array1=array[0].split('.NASDQ')
26 |                 market='NASDQ'
27 |                 name=array1[1].strip()
28 |                 code=array1[0].strip()
29 |             if('.NYSE' in array[0]):
30 |                 array1=array[0].split('.NYSE')
31 |                 market='NYSE'
32 |                 name=array1[1].strip()
33 |                 code=array1[0].strip()
34 |             if('.AMEX' in array[0]):
35 |                 array1=array[0].split('.AMEX')
36 |                 market='AMEX'
37 |                 name=array1[1].strip()
38 |                 code=array1[0].strip()
39 |             
40 |             out.write(name+'\t'+code+'\t'+market+'\n')
41 |     out.close()
42 |     
43 | if __name__ == '__main__':
44 |     china()
45 |     american()


--------------------------------------------------------------------------------
/data_download.py:
--------------------------------------------------------------------------------
  1 | # -*- coding: utf-8 -*- #
  2 | import urllib 
  3 | import urllib2 
  4 | import requests
  5 | import codecs
  6 | import socket
  7 | 
  8 | def download():
  9 |     with codecs.open('D:\\programing\\Python\\MLHomework\\china_codeName.txt','r','utf-8') as f:
 10 |         for line in f:
 11 |             array=line.strip().split('\t')
 12 |             name=array[0]
 13 |             code=array[1]
 14 |             market=array[2]
 15 |             if(u'上证'==market):
 16 |                 print code+u'上证'
 17 |                 url='http://table.finance.yahoo.com/table.csv?s='+code+'.ss'
 18 |                 urllib.urlretrieve(url, 'D:\\programing\\Python\\MLHomework\\stockData_china\\'+code+'.ss.csv')
 19 |             elif(u'深成'==market):
 20 |                 print code+u'深成'
 21 |                 url='http://table.finance.yahoo.com/table.csv?s='+code+'.sz'
 22 |                 urllib.urlretrieve(url, 'D:\\programing\\Python\\MLHomework\\stockData_china\\'+code+'.sz.csv')
 23 | 
 24 | def american_download():
 25 |     #socket.setdefaulttimeout(15)
 26 |     n=0
 27 |     with codecs.open('I:\\MLHomework\\american_codeName.txt','r','utf-8') as f:
 28 |         for line in f:
 29 |             array=line.strip().split('\t')
 30 |             name=array[0]
 31 |             code=array[1]
 32 |             market=array[2]
 33 | 
 34 |             n+=1
 35 |             print code,n
 36 |             url='http://table.finance.yahoo.com/table.csv?s='+code+'&d=8&e=24&f=2015&g=d&a=1&b=1&c=2010&ignore=.csv'
 37 |             try:
 38 |                 urllib.urlretrieve(url, 'I:\\MLHomework\\stockData_american1\\'+code+'.csv')
 39 |             except:
 40 |                 print 'error'
 41 | 
 42 | 
 43 | def china_download_day(seqFilePath,downloadDir,outDataPath,y,m,d):
 44 |     data=[]
 45 |     out=codecs.open(outDataPath,'w','utf-8')#输出data数据
 46 |     n=0
 47 |     with codecs.open(seqFilePath,'r','utf-8') as f:
 48 |         for line in f:
 49 |             array=line.strip().split('.')
 50 |             code=array[0]
 51 |             market=array[1]
 52 |             print(code+market+'\t'+str(n))
 53 |             n+=1
 54 |             savedFilePath=''
 55 |             if(u'ss'==market):
 56 |                 url='http://table.finance.yahoo.com/table.csv?s='+code+'.ss&d='+str(m-1)+'&e='+str(d)+'&f='+str(y)+'&g=d&a='+str(m-1)+'&b='+str(d)+'&c='+str(y)+'&ignore=.csv'
 57 |                 urllib.urlretrieve(url, downloadDir+'\\'+code+'.ss.csv')
 58 |                 savedFilePath=downloadDir+'\\'+code+'.ss.csv'
 59 |             elif(u'sz'==market):
 60 |                 url='http://table.finance.yahoo.com/table.csv?s='+code+'.sz&d='+str(m-1)+'&e='+str(d)+'&f='+str(y)+'&g=d&a='+str(m-1)+'&b='+str(d)+'&c='+str(y)+'&ignore=.csv'
 61 |                 urllib.urlretrieve(url, downloadDir+'\\'+code+'.sz.csv')
 62 |                 savedFilePath=downloadDir+'\\'+code+'.sz.csv'
 63 |             #下载完成之后就可以得到这个文件中的数据       
 64 |             with codecs.open(savedFilePath,'r','utf-8') as f:
 65 |                 firstline=True
 66 |                 for line in f:
 67 |                     if(firstline):
 68 |                         firstline=False
 69 |                         continue
 70 |                     else:#那些error的会出错 最后break就行
 71 |                         rise=1#上涨
 72 |                         date=''
 73 |                         try:
 74 |                             array=line.strip().split(',')
 75 |                             date=array[0]
 76 |                             openValue=float(array[1])
 77 |                             closeValue=float(array[4])
 78 |                             if(closeValue<openValue):
 79 |                                 rise=0#下跌
 80 |                         except:
 81 |                             rise=0#下跌
 82 |                             print('error')
 83 |                         #组个日期字符串出来
 84 |                         dateT=str(y)+'-'
 85 |                         if(m<10):
 86 |                             dateT+='0'+str(m)+'-'
 87 |                         else:
 88 |                             dateT+=str(m)+'-'
 89 |                     
 90 |                         if(d<10):
 91 |                             dateT+='0'+str(d)
 92 |                         else:
 93 |                             dateT+=str(d)
 94 |                             
 95 |                         #如果日期和文件中的相同那就ok
 96 |                         if(dateT==date):
 97 |                             out.write(str(rise)+'\t')
 98 |                         else:
 99 |                             out.write(str(0)+'\t')
100 |                         break
101 |     out.close()
102 |     
103 | if __name__ == '__main__':
104 |     #download()
105 |     american_download()
106 |     #china_download_day('I:\\MLHomework\\china_sequence.txt','I:\\MLHomework\\stockData_china_1day','I:\\MLHomework\\1dayData.txt',2015,8,26)


--------------------------------------------------------------------------------
/getData/__init__.py:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/jxls080511/stockPrediction/7ccb30b9f2b3537347e00b613ec0dbbaa97c6bc7/getData/__init__.py


--------------------------------------------------------------------------------
/getData/getTrainData.py:
--------------------------------------------------------------------------------
 1 | import os
 2 | #起止年月，end年份月份不包含，usDir是美股数据的文件夹位置，outDir是输出（训练/测试）文件的位置
 3 | def getData(startY,endY,startM,endM,usDir,outDir):    
 4 |     record_dict={}
 5 |     for filename in os.listdir('I:\\MLHomework\\stockData_china'):
 6 |         print(filename)
 7 |         stock_dict={}
 8 |         with open('I:\\MLHomework\\stockData_china\\'+filename,'r',encoding='utf-8') as f:
 9 |             firstline=True
10 |             for line in f:
11 |                 if(firstline):
12 |                     firstline=False
13 |                     continue
14 |                 else:
15 |                     array=line.strip().split(',')
16 |                     date=array[0]
17 |                     openValue=float(array[1])
18 |                     closeValue=float(array[4])
19 |                     rise=1#上涨
20 |                     if(closeValue<=openValue):
21 |                         rise=0#下跌
22 |                     stock_dict[date]=rise
23 |         record_dict[filename]=stock_dict
24 |     
25 | 
26 |     for us_filename in os.listdir(usDir):
27 |         print('us'+us_filename)
28 |         target_dict={}
29 |         with open(usDir+'\\'+us_filename,'r',encoding='utf-8') as f:
30 |             firstline=True
31 |             for line in f:
32 |                 if(firstline):
33 |                     firstline=False
34 |                     continue
35 |                 else:
36 |                     array=line.strip().split(',')
37 |                     date=array[0]
38 |                     openValue=float(array[1])
39 |                     closeValue=float(array[4])
40 |                     rise=1#上涨
41 |                     if(closeValue<openValue):
42 |                         rise=0#下跌
43 |                     target_dict[date]=rise
44 |     
45 |         out=open(outDir+'\\'+us_filename+'.data','w',encoding='utf-8')
46 | #         out2=open('I:\\MLHomework\\china_sequence.txt','w',encoding='utf-8')
47 | #         first=True
48 |         #用2015-1-1到2015-4-30的数据作为训练数据
49 |         for year in range(startY,endY):
50 |             for month in range(startM,endM):
51 |                 for day in range(1,32):
52 |                     date=str(year)+'-'
53 |                     if(month<10):
54 |                         date+='0'+str(month)+'-'
55 |                     else:
56 |                         date+=str(month)+'-'
57 |                     
58 |                     if(day<10):
59 |                         date+='0'+str(day)
60 |                     else:
61 |                         date+=str(day)   
62 |                 
63 |                     #遍历所有中国股票
64 |                     if(date in target_dict.keys()):
65 |                         out.write(date+'\t')
66 |                         for stockID in record_dict.keys():
67 |                             #我需要中国股票作为训练数据的顺序输出成文件给我
68 | #                             if(first):
69 | #                                 out2.write(stockID[0:len(stockID)-4]+'\n')
70 | #                             else:
71 | #                                 out2.close()
72 |                             stock_dict=record_dict[stockID]
73 |                             if(date in stock_dict.keys()):
74 |                                 out.write(str(stock_dict[date])+'\t')
75 |                             else:
76 |                                 out.write('0\t')
77 |                         first=False            
78 |                         #把这个股票的涨跌当做结果写在最后
79 |                         out.write(str(target_dict[date])+'\n')
80 | 
81 |         out.close()
82 |                       
83 | if __name__ == '__main__':
84 |     getData(2015,2016,5,9,'I:\\MLHomework\\stockData_american','I:\\MLHomework\\Data\\4\\testData4m')


--------------------------------------------------------------------------------
/美股涨跌预测系统的探究-141499罗斌.docx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/jxls080511/stockPrediction/7ccb30b9f2b3537347e00b613ec0dbbaa97c6bc7/美股涨跌预测系统的探究-141499罗斌.docx


--------------------------------------------------------------------------------