def main():
var = '' #just want update to repeat not entire thing
while var != 'null':
company_name = raw_input(
"Enter the name of the company you're searching for "
) #THis asks for a company name
var = get_companysymbol(
company_name
) #this searchs the company list we have and returns null if not found
var = company_name
if var != 'null': # ONLY PRECED IF WE HAVE A COMPANY
timeBegin = int(
input('Enter Start year ')
) - 2 #Gets the start date and pushes it back 2 years to show enough data to fit the screen nicely
totalDataCurrent = Fetching.fetchDataToday(
var, timeBegin
) # This gets all the data from the start year to 3 days ago (give or take a work day)
googData = Fetching.fetchGoogData(
var) # Fetch Todays data from google finance
Prediction_Data = Fetching.fetchDataSpec(
var, (datetime.now() + timedelta(days=-45))
) # Get the data from just the past month for the prediciton part
Prediction_Data_Length = len(
Prediction_Data.High
) # Lenght of the predictin Data to save the recalc of it
Coeffcients = LinearAlgebra.coeffcients_Generator(
LinearAlgebra.makeXVals_Matrix(10, timeBegin,
Prediction_Data_Length),
LinearAlgebra.makeY_Matrix(Prediction_Data.High)
) #coeffcients for prediction fucntion a0-a10
Prediction_Model = LinearAlgebra.makeOutY(
Coeffcients, Prediction_Data_Length, timeBegin,
totalDataCurrent.High, googData
) # Gets Prediciton Model or scatter of predicted points these points are also normalized
pointY = LinearAlgebra.getPointY(
Coeffcients, timeBegin, googData[0]['LastTradePrice'],
totalDataCurrent.High[len(totalDataCurrent) - 1]
) #gets Predictiion Point for the next day independently so I can calculate individual days
R = RSI.PredictRSI(totalDataCurrent.Close)
Graphing.totalTogether(
var, totalDataCurrent, googData, Prediction_Model, pointY,
R) #Print Final Graph with everything together
pass
def getGraph():
stockName = request.args.get('s')
var=stockName
old_stdout = sys.stdout
sys.stdout = open(os.devnull, "w")
if var != 'null': # ONLY PRECEDE IF WE HAVE A COMPANY
Coefficients = numpy.zeros(shape =(11,1))
timeBegin=2010
totalDataCurrent=Fetching.fetchDataToday(var,timeBegin) # This gets all the data from the start year to 3 days ago (give or take a work day)
googData=Fetching.fetchGoogData(var) # Fetch Todays data from google finance
Prediction_Data=Fetching.fetchDataSpec(var,(datetime.now()+timedelta(days=-45))) # Get the data from just the past month for the prediciton part
Prediction_Data_Length=len(Prediction_Data.Close) # Lenght of the predictin Data to save the recalc of it
#if Cache2.Search(var) ==0:
Coefficients=LinearAlgebra.coefficients_Generator(LinearAlgebra.makeXVals_Matrix(10,timeBegin,Prediction_Data_Length),LinearAlgebra.makeY_Matrix(Prediction_Data.Low)) #coeffcients for prediction fucntion a0-a10
#Cache2.Cache_Predictions(var,Coefficients) # after calculating store the data in cache
#else:
#Coefficients = Cache2.Fetch_Cache(var) # fetch from cache if the company data is stored and it's recent ( less than 3 days from prediction)
Prediction_Model=LinearAlgebra.makeOutY(Coefficients,Prediction_Data_Length,timeBegin,totalDataCurrent.High,googData) # Gets Prediciton Model or scatter of predicted points these points are also normalized
pointY=LinearAlgebra.getPointY(Coefficients,timeBegin,googData[0]['LastTradePrice'],totalDataCurrent.High[len(totalDataCurrent)-1]) #gets Predictiion Point for the next day independently so I can calculate individual days
R=RSI.PredictRSI(totalDataCurrent.Close)
pointY=LinearAlgebra.getPointY(Coefficients,timeBegin,googData[0]['LastTradePrice'],totalDataCurrent.Low[len(totalDataCurrent)-1]) #gets Predictiion Point for the next day independently so I can calculate individual days
url=Graphing.totalTogether(var,totalDataCurrent,googData,Prediction_Model,pointY,R) #Print Final Graph with everything together
url = url + ".embed?width=640&height=480"
ret = "<iframe width='640' height='480' frameborder='0' scrolling='no' src='" + url + "'> </iframe>"
sys.stdout.close()
sys.stdout = old_stdout
return ret
else:
sys.stdout.close()
sys.stdout = old_stdout
return "false"
def analyze(stock_code, date):
total_score = 0.0
total_score += BBI.analyze(stock_code, date, BBI_PARM_PERIOD_TRACE_FOR_HIGH, BBI_PARM_PERIOD_TRACE_HIS)[0]
total_score += BIAS.analyze(stock_code, date)[0]
total_score += BOLL.analyze(stock_code, date, BOLL_PARM_PERIOD)[0]
total_score += BRAR.analyze(stock_code, date, BRAR_PARM_PERIOD_HIS)[0]
total_score += CCI.analyze(stock_code, date, CCI_PARM_PERIOD)[0]
total_score += DMA.analyze(stock_code, date, DMA_PARM_PERIOD)[0]
total_score += DMI.analyze(stock_code, date, DMI_PARM_PERIOD)[0]
total_score += KDJ.analyze(stock_code, date)[0]
total_score += MA.analyze(stock_code, date, MA_PARM_PERIOD)[0]
total_score += OBV.analyze(stock_code, date, OBV_PARM_PERIOD_TRACE_HIS)[0]
total_score += PSY.analyze(stock_code, date)[0]
total_score += ROC.analyze(stock_code, date, ROC_PARM_PERIOD_TRACE_HIS)[0]
total_score += RSI.analyze(stock_code, date, RSI_PARM_PERIOD_TRACE_HIS)[0]
total_score += VR.analyze(stock_code, date)[0]
total_score += WR.analyze(stock_code, date, WR_PARM_PERIOD_TRACE_HIS)[0]
bd.updateAnalysisData("TOTAL_SCORE", str(total_score), stock_code, date)
def latestScores(start_date, end_date, tickerSymbol):
debug = 0
if debug == 1:
print("start_date, end_Date, tickersymbol")
print(start_date)
print(end_date)
print(tickerSymbol)
url = "https://raw.githubusercontent.com/datasets/covid-19/master/data/time-series-19-covid-combined.csv"
filename = url
df = pd.read_csv(filename, lineterminator='\n')
tdate = start_date
sDates, sCloses = Ticker.Ticker(start_date, end_date, tickerSymbol)
# Check if Ticker is Valid and if not use default tickersymbol
validTicker = True
if len(sDates) < 1:
validTicker = False
sDates, sCloses = Ticker.Ticker(start_date, end_date, defaultTS)
scDates, tCases, scCloses = CV_Stats.CV_Stats(tdate, sDates, sCloses, df,
end_date)
sDates = scDates
sCloses = scCloses
sDates1 = pd.to_datetime(sDates)
sDates1 = sDates1.strftime("%m-%d")
score = Correlation.Correlate(sCloses, tCases)
if debug == 1:
print("sDates")
print(sDates)
score1 = RSI.computeRSI(sCloses, sDates, 10, 30)
return score, score1, validTicker, sDates1, sCloses, tCases
def main():
p_f_st = 0
p_f_no = 0
p_f_k = 0
p_f_h = 0
#---------------------------------------------------------------------
#AntTreeアルゴリズムを用いた(株価データ)クラスタリングシステム
#--STOCH -> 確率論的アルゴリズム
#--NO_THRESHOLDS -> 決定論的アルゴリズム
#---- Memo -----------------------------------------------------------
#--K-means法は比較用に作成
#--株価はyahoo! financeから2015年4/1～11/30までの営業日で取得
#--12/8追記 4/1～11/30(163次元)でメモリエラー発生 -> クラスタリングできていない
#--1/12 4/1～11/30(163次元)できた！！（家PC）
#--1/12 クラスタリング結果の評価関数(Pseudo_F.py)追加
#---------------------------------------------------------------------
# -*- coding:utf-8 -*-
import datetime
import pandas as pd
from pandas import Series, DataFrame
import matplotlib.pyplot as plt
#自作モジュールのインポート
import STOCH_StockPrice as st_sp
import NO_THRESHOLDS_StockPrice as nt_sp
import K_means_StockPrice as kmeans_sp
import Hierarchy_clustering as h_c
import cluster_Ant as cl_ant
import Return_Index as r_index
import RSI as rsi
import Move_Average as m_ave
import Pseudo_F as p_f
#date 4/1～11/30まで
#4/1 ~ 5/18 : 30日分
#4/1 ~ 6/29 : 60日分
#4/1 ~ 8/11 : 90日分
#4/1 ~ 9/27 : 120日分
start_date = datetime.date(2015,4,1)
end_date = datetime.date(2015,9,27)
#企業コードと企業名
companies_test = {9682:'DTS', 9742:'アイネス', 9613:'NTTデータ', 2327:'新日鉄住金ソリューションズ',
9640:'セゾン情報システムズ', 3626:'ITホールディングス', 2317:'システナ',
4684:'オービック', 9739:'NSW', 4726:'ソフトバンク・テクノロジー', 4307:'野村総合研究所',
9719:'SCSK', 4793:'富士通ビー・エス・シー', 4812:'ISID', 8056:'日本ユニシス'}
companies_it = {9682:'DTS', 9742:'アイネス', 9613:'NTTデータ', 2327:'新日鉄住金ソリューションズ',
3626:'ITホールディングス', 2317:'システナ',
4684:'オービック', 9739:'NSW', 4726:'ソフトバンク・テクノロジー', 4307:'野村総合研究所',
9719:'SCSK', 4812:'ISID', 8056:'日本ユニシス',
3765:'ガンホー・オンライン・エンターテイメント', 4689:'ヤフー', 4716:'日本オラクル',
4739:'伊藤忠テクノソリューションズ', 4825:'ウェザーニュース', 7844:'マーベラス', 9432:'日本電信電話',
9433:'KDDI', 9437:'NTTドコモ', 9697:'カプコン', 9766:'コナミ', 9984:'ソフトバンクグループ',
9749:'富士ソフト', 3632:'グリー', 3656:'KLab'} #情報通信系
companies_food = {2001:'日本製粉', 2002:'日清製粉グループ本社', 2003:'日東富士製粉', 2109:'三井製糖', 2201:'森永製菓',
2206:'江崎グリコ', 2211:'不二家', 2212:'山崎製パン', 2217:'モロゾフ', 2220:'亀田製菓', 2229:'カルビー',
2264:'森永乳業', 2267:'ヤクルト本社', 2270:'雪印メグミルク', 2281:'プリマハム', 2282:'日本ハム',
2284:'伊藤ハム', 2288:'丸大食品', 2501:'サッポロホールディングス', 2502:'アサヒホールディングス', 2503:'キリンホールディングス',
2579:'コカ・コーラウエスト', 2593:'伊藤園'} #食料品
companies_retail = {2651:'ローソン', 2681:'ゲオホールディングス', 2730:'エディオン', 3048:'ビックカメラ', 3197:'すかいらーく',
3382:'セブン＆アイホールディングス', 7550:'ゼンショーホールディングス', 7581:'サイゼリヤ', 7630:'壱番屋',
8028:'ファミリーマート', 8175:'ベスト電器', 8179:'ロイヤルホールディングス', 8200:'リンガーハット',
9831:'ヤマダ電器'} #小売
companies_device = {4902:'コニカミノルタ', 6448:'ブラザー工業', 6501:'日立製作所', 6502:'東芝', 6503:'三菱電機',
6506:'安川電機', 6516:'三洋電機', 6645:'オムロン', 6701:'NEC', 6702:'富士通', 6703:'OKI',
6724:'セイコーエプソン', 6752:'パナソニック', 6753:'シャープ', 6758:'ソニー', 6762:'TDK',
6981:'村田製作所', 7751:'キャノン', 7752:'リコー', 8035:'東京エレクトロン'} #電気機器
companies_service = {2193:'クックパッド', 2371:'カカクコム', 2379:'ディップ', 2433:'博報堂DIYホールディングス',
2432:'ディー・エヌ・エー', 4324:'電通', 4751:'サイバーエージェント', 4755:'楽天'} #サービス
companies_mix = {9437:'NTTドコモ', 4307:'野村総合研究所', 2327:'新日鉄住金ソリューションズ', 8056:'日本ユニシス', 4812:'ISID',
2503:'キリンホールディングス', 2579:'コカ・コーラウエスト', 2288:'丸大食品', 2109:'三井製糖',
2730:'エディオン', 7550:'ゼンショーホールディングス', 7630:'壱番屋', 8175:'ベスト電器', 8179:'ロイヤルホールディングス',
6502:'東芝', 6702:'富士通', 8035:'東京エレクトロン', 6753:'シャープ',
2432:'ディー・エヌ・エー', 4324:'電通', 4751:'サイバーエージェント', 4755:'楽天', 2193:'クックパッド'} #混合
#↑printでの表示は工夫が必要... とりあえず使いたいのはkeyだけ
companies = companies_service #クラスタリングする業種
#update → ディクショナリの連結
#companies.update(companies_food)
#companies.update(companies_retail)
#companies.update(companies_device)
#companies.update(companies_service)
#print 'companies: ',len(companies)
#print 'term : '+str(start_date)+' -- '+str(end_date)
#start_dateからend_dateまでの期間のリターンインデックスを計算，csvファイルで保存
#戻り値は営業日のdatetimeオブジェクト
#dates = r_index.main(start_date, end_date, companies)
#fname1 = 'return_index_values.csv'
#fname2 = 'return_index_codes.csv'
#start_dateからend_dateまでのRSIを計算，csvファイルで保存
#戻り値はdatetimeオブジェクト
n = 10 #10日間でのRSI
dates = rsi.main(start_date, end_date, n, companies)
fname1 = 'RSI_values.csv'
fname2 = 'RSI_codes.csv'
#print 'dates : ', len(dates)
#--- STOCH_StockPrice ---
Ant, X, count = st_sp.main(0.99, 0.1, fname1, fname2) #引数1:alpha1, 引数2:alpha2
label = cl_ant.ant_label(Ant)
tmp = []
label_max = max(label) #クラスタ数
for i in range(label_max+1):
tmp.append({})
c = 0 #counter
codes = companies.keys()
#グラフ描画の準備
for i in label:
for j in range(label_max+1):
if j == i:
tmp[i][codes[c]] = X[c]
else : pass
c = c+1
PseudoF = p_f.main(tmp, label_max+1)
p_f_st = PseudoF
k = label_max+1 #k-meansのクラスタ数
#--- NO_THRESHOLDS_StockPrice ---
Ant, X, count = nt_sp.main(fname1, fname2)
label = cl_ant.ant_label(Ant)
tmp = []
label_max = max(label) #クラスタ数
for i in range(label_max+1):
tmp.append({})
c = 0 #counter
codes = companies.keys()
#グラフ描画の準備
for i in label:
for j in range(label_max+1):
if j == i:
tmp[i][codes[c]] = X[c]
else : pass
c = c+1
PseudoF = p_f.main(tmp, label_max+1)
p_f_no = PseudoF
#--- K-means ---
label, X = kmeans_sp.main(fname1, k) #kはstochと同じに指定
tmp = []
label_max = max(label)
for i in range(label_max+1):
tmp.append({})
c = 0
codes = companies.keys()
#グラフ描画の準備
for i in label:
for j in range(label_max+1):
if j == i:
tmp[i][codes[c]] = X[c]
else : pass
c = c+1
PseudoF = p_f.main(tmp, label_max+1)
p_f_k = PseudoF
#--- 階層的クラスタリング---
#print 'Hierarchy Clustering'
tmp = h_c.main(fname1, fname2, 'single')
PseudoF = p_f.main(tmp, len(tmp))
p_f_h = PseudoF
return p_f_st, p_f_no, p_f_k, p_f_h
def pairs(cusip1, cusip2, dt):
cols = [
'Stock1', 'Stock2', 'YYYYMM', 'SSD_P', 'MHD_P', 'Std2', 'RTN', 'CNT',
'DAYS', 'SSD_V', 'MHD_V', 'SSD_T', 'MHD_T', 'SSD_R', 'MHD_R'
]
pos_pairs = pd.DataFrame(columns=cols)
im_cols = [
'DATE', 'YYYYMM', 'PRC', 'SHROUT', 'DIVAMT', 'OPENPRC', 'ASKHI',
'BIDLO', 'VOL'
]
a = dt[im_cols][dt.CUSIP == cusip1]
b = dt[im_cols][dt.CUSIP == cusip2]
strt = max(min(a.DATE), min(b.DATE))
end = min(max(a.DATE), max(b.DATE))
a = a[(a.DATE >= strt) & (a.DATE <= end)]
b = b[(b.DATE >= strt) & (b.DATE <= end)]
a = a.sort_values(['DATE']).reset_index(drop=True)
b = b.sort_values(['DATE']).reset_index(drop=True)
a['RTN'] = daily_return(a)
b['RTN'] = daily_return(b)
a['Vola'] = Dua.VE(a)
b['Vola'] = Dua.VE(b)
a['TNOV'] = Tao.turnover(a)
b['TNOV'] = Tao.turnover(b)
a['RSI'] = Lau.get_rsi(a)
b['RSI'] = Lau.get_rsi(b)
#pdb.set_trace()
merge_col = ['DATE', 'YYYYMM', 'PRC', 'RTN', 'Vola', 'VOL', 'TNOV', 'RSI']
c = pd.merge(a[merge_col], b[merge_col], on=['DATE', 'YYYYMM'])
c = c.fillna(0)
#pdb.set_trace()
mt = sorted(list(set(c.YYYYMM)))
for mn in mt[11:len(mt) - 6]:
d = c[c.YYYYMM <= mn]
e = c[(c.YYYYMM > mn)
& (c.YYYYMM <= str(int(mn[0:4]) + int((6 + int(mn[4:6])) / 12)) +
str((6 + int(mn[4:6])) % 12).zfill(2))]
std2 = std_of_his_sprd(d.PRC_x, d.PRC_y)
rtn, cnt, days = pair_trading(e, std2)
pos_pairs = pos_pairs.append(
{
'Stock1':
cusip1,
'Stock2':
cusip2,
'YYYYMM':
mn,
'SSD_P':
sum_sqr_diff_prc(d.PRC_x, d.PRC_y),
'MHD_P':
sum_mah_diff_prc(d.PRC_x, d.PRC_y),
'Std2':
std2,
'RTN':
rtn,
'CNT':
cnt,
'DAYS':
days,
'SSD_V':
Dua.ssd_v(d.Vola_x, d.Vola_y),
'MHD_V':
Dua.smd_v(d.Vola_x, d.Vola_y),
'SSD_T':
Tao.ssd_t(d.TNOV_x, d.TNOV_y),
'MHD_T':
Tao.smd_t(d.TNOV_x, d.TNOV_y),
'SSD_R':
Lau.get_sum_of_square_difference_of_rsi(d, d.RSI_x, d.RSI_y),
'MHD_R':
Lau.get_maha_distance_of_rsi(d, d.RSI_x, d.RSI_y)
},
ignore_index=True)
return (pos_pairs)
'''
Created on Jun 15, 2017
@author: michaelstecklein
'''
import Database
import RSI
import StockData
import matplotlib.pyplot as plt
#Database.rsi_add_default_entries()
rsis = RSI.get_RSIs(Database.get_stock("AAPL"), StockData.createSDate("2017-01-01"), StockData.createSDate("2017-06-22"))
plt.plot(rsis)
plt.plot((0, len(rsis)), (70, 70), 'r-')
plt.plot((0, len(rsis)), (30, 30), 'g-')
plt.ylabel('RSI')
plt.show()
print "\ndone"
import DataLoading
import MACD
import RSI
f = open("idx.txt","r")
fw = open("macd_histogram.csv","w")
fw.write("Ticker,macd_histogram,rsi,volume\n")
for ticker in f.readlines():
data = DataLoading.getData(ticker.strip(), 700)
df = MACD.MACD(data)
macd_histogram = df['MACD Histogram'].iat[-1]
#if MACD > signal line then buy else sell
macd = df['MACD'].iat[-1] > df['Signal Line'].iat[-1]
df = RSI.RSI(data)
rsi = df['rsi'].iat[-1]
volume = df['Volume'].iat[-1]
if(macd_histogram <= 2 and rsi <= 30 and macd):
fw.write(ticker.strip() + "," + str(macd_histogram) + "," + str(rsi) + "," + str(volume) + "\n")
#print(ticker.strip(), " = ", macd_histogram)
f.close()
fw.close()
#companies.update(companies_retail)
#companies.update(companies_device)
#companies.update(companies_service)
print 'companies: ',len(companies)
print 'term : '+str(start_date)+' -- '+str(end_date)
#start_dateからend_dateまでの期間のリターンインデックスを計算，csvファイルで保存
#戻り値は営業日のdatetimeオブジェクト
#dates = r_index.main(start_date, end_date, companies)
#fname1 = 'return_index_values.csv'
#fname2 = 'return_index_codes.csv'
#start_dateからend_dateまでのRSIを計算，csvファイルで保存
#戻り値はdatetimeオブジェクト
n = 10 #10日間でのRSI
dates = rsi.main(start_date, end_date, n, companies)
fname1 = 'RSI_values.csv'
fname2 = 'RSI_codes.csv'
print 'dates : ', len(dates)
#--- STOCH_StockPrice ---
Ant, X, count = st_sp.main(0.99, 0.1, fname1, fname2) #引数1:alpha1, 引数2:alpha2
label = cl_ant.ant_label(Ant)
tmp = []
label_max = max(label) #クラスタ数
for i in range(label_max+1):
tmp.append({})
c = 0 #counter
import RSI
coin = input("Input coin>>")
RSI.getRSI()
print("The RSI of",coin,"is",RSI)
if RSI>80:
print("You should sell")
elif RSI<20:
print("You should buy")
import Datasheet_Download
import preprocessing
import RSI
toDownload = input("Do you want to downlaod the historical data (y/n) :")
toContinue = "y"
if (toDownload.lower() == "y"):
Datasheet_Download.Download_Dataheet()
toContinue = input("Do you want to continue(y/n) : ")
if (toContinue == "y"):
preprocessing.preprocess()
RSI.Execute()
def run(self): # 把要执行的代码写到run函数里面 线程在创建后会直接运行run函数
print("Starting " + self.name)
print
"Exiting2 " + self.name
while True:
code = 'sh.000001'
today = datetime.datetime.now().strftime('%Y%m%d')
startdate = '2017-01-01'
enddate = f'{today[0:4]}-{today[4:6]}-{today[6:8]}'
# 获取当前时间
now_time = datetime.datetime.now()
period = 1
Vol = vol.computeVOL(code, startdate, enddate, period)
print(Vol[-1])
# rsi_6days = rsi.computeRSI(code, startdate, enddate)
# print(rsi_6days[-1])
# imnow = time.strftime("%Y-%m-%d %H:%M:%S")
# print(imnow)
# time.sleep(1)
# print(now_time.time().hour)
# print(now_time.time().minute)# 创建新线程
print
"Exiting Main Thread"
# print(now_time.time().second)
# print(now_time.time().hour+":"+now_time.time().second +":"+now_time.time().second())
# 系统时间XX点XX分XX秒，触发策略启动
if now_time.time().hour == 9 and now_time.time(
).minute == 37 and now_time.time().second == 00:
#print('启动KDJ策略')
# ##########KDJ策略体###########
# #enddate不是交易日，测试会出现接收数据异常 list index out of range
# # J = kdj.computeKDJ(code, startdate, enddate)
# # 买入测试日期
# #J = kdj.computeKDJ(code, startdate,'2019-09-26')
# #卖出测试日期
# J = kdj.computeKDJ(code, startdate,'2019-08-21')
# print(J[-1])
#
# if J[-1] < 0:
# # work.button_start(QWidget)
# print('启动买入策略')
# autoTrade.buy(159928)
#
# print("交易成功")
# ######################
# if J[-1] > 100:
# # work.button_start(QWidget)
# print('启动卖出策略')
# autoTrade.sell(159928)
#
#
# print("交易成功")
# ######################
print('启动RSI策略')
##########RSI策略体###########
# enddate不是交易日，测试会出现接收数据异常 list index out of range
#rsi_6days = rsi.computeRSI(code, startdate, enddate)
# 买入测试日期
# J = rsi.computeRSI(code, startdate,'2019-09-26')
# 卖出测试日期
rsi_6days = rsi.computeRSI(code, startdate, '2019-09-09')
print(rsi_6days[-1])
if rsi_6days[-1] < 20:
# work.button_start(QWidget)
print('启动买入策略')
autoTrade.buy(159928)
print("交易成功")
######################
if rsi_6days[-1] > 80:
# work.button_start(QWidget)
print('启动卖出策略')
autoTrade.sell(159928)
print("交易成功")
def update_indicators():
RSI.update_RSI()
