def getData(self, tickers) :
settings = Settings()
dp = TimeSeries(settings).get_agg_ETF_data(tickers)
dp = dp.fillna(method='pad', axis=0)
dp = dp.fillna(method='bfill', axis=0)
dp = dp.dropna()
dp = dp.reindex_axis(['open', 'high', 'low', 'close', 'volume', 'price'], axis=2)
return dp
import Data.StockDataManager as sm
import Data.TimeSeries as ts
settings = sm.Settings()
tmgr = sm.TaskManager(settings)
data = ts.TimeSeries(settings).get_stock_series(['000001'],
start='2012-01-01',
fields=['close'])
price = data['close']
import pyhht.emd as emd
import numpy as np
imfs = emd.EMD(np.array(price))
result = ts.adfuller(x)
# Calculate the half-life of reversion
price = pd.Series(x)
lagged_price = price.shift(1).fillna(method="bfill")
delta = price - lagged_price
beta = np.polyfit(lagged_price, delta,
1)[0] #Use price(t-1) to predicate delta.
half_life = (-1 * np.log(2) / beta)
return result[0], result[1], half_life
from Data.StockDataManager import *
from Data.TimeSeries import *
if __name__ == "__main__":
tickers = ['GOOG/NYSE_SPY']
settings = Settings()
dp = TimeSeries(settings).get_agg_ETF_data(tickers)
df = dp[:, :, 'price']
#df.plot(figsize=[20,12])
n = df.copy()
# n['scaling'] = scaling(df, 50)
# n['centering'] = centering(df, 50)
n['normalization'] = Normalization(df.values, 50)
n['regnormal'] = RegularNormalization(df.values, 50)
n.plot()
print 'done!'
'GOOG/NYSE_VWO', # Vanguard Emerging Market Stock ETF
#'GOOG/NYSE_VNQ', # Vanguard MSCI US Reits
'GOOG/NYSE_IYR', # iShares U.S. Real Estate ETF
'GOOG/NYSE_RWX', # SPDR DJ Wilshire Intl Real Estate ETF
'GOOG/NYSEARCA_TLT', # 20 Years Treasury ETF
'GOOG/NYSEARCA_TLH', # 15-20 Years Treasury
'GOOG/AMEX_GSG', # GSCI Commodity-Indexed Trust Fund
'GOOG/NYSEARCA_GLD' # SPDR Gold ETF
]
settings = Settings()
dp = TimeSeries(settings).get_agg_data(tickers)
dp = dp.fillna(method='pad', axis=0)
dp = dp.fillna(method='bfill', axis=0)
dp = dp[:,:,:]
dp[:, :,'price'].plot(figsize=[20,10])
df_rets = dp[:,:,'price'].pct_change().dropna()
'GOOG/NYSEARCA_XLI',
'GOOG/NYSEARCA_XLK',
'GOOG/NYSEARCA_XLP',
'GOOG/NYSEARCA_XLU',
'GOOG/NYSEARCA_XLV',
'GOOG/NYSEARCA_XLY'
]
bm_tickers = [
'GOOG/NYSE_SPY',
'GOOG/NYSEARCA_TLT',
'GOOG/NYSEARCA_GLD' #gold
]
settings = Settings()
dp = TimeSeries(settings).get_agg_data(sector_tickers+bm_tickers)
dp = dp.fillna(method='pad', axis=0)
dp = dp.fillna(method='bfill', axis=0)
dp = dp[:,'2000-01-01'::,:]
dp = dp.dropna()
rets = pd.DataFrame()
rets[1] = Momentum_Sector_Rotation(dp).run_trading().portfolio_value
#rets[1] = Portfolio1(dp).run_trading().portfolio_value
#rets[2] = Portfolio2(dp).run_trading().portfolio_value
# rets[3] = Portfolio3(dp).run_trading().portfolio_value
# rets[4] = Portfolio4(dp).run_trading().portfolio_value
# rets[5] = Portfolio5(dp).run_trading().portfolio_value
#
]
sector_tickers = [
'GOOG/NYSEARCA_XLB', 'GOOG/NYSEARCA_XLE', 'GOOG/NYSEARCA_XLF',
'GOOG/NYSEARCA_XLI', 'GOOG/NYSEARCA_XLK', 'GOOG/NYSEARCA_XLP',
'GOOG/NYSEARCA_XLU', 'GOOG/NYSEARCA_XLV', 'GOOG/NYSEARCA_XLY'
]
bm_tickers = [
'GOOG/NYSE_SPY',
'GOOG/NYSEARCA_TLT',
'GOOG/NYSEARCA_GLD' #gold
]
settings = Settings()
dp = TimeSeries(settings).get_agg_data(sector_tickers + bm_tickers)
dp = dp.fillna(method='pad', axis=0)
dp = dp.fillna(method='bfill', axis=0)
dp = dp[:, '2000-01-01'::, :]
dp = dp.dropna()
rets = pd.DataFrame()
rets[1] = Momentum_Sector_Rotation(dp).run_trading().portfolio_value
#rets[1] = Portfolio1(dp).run_trading().portfolio_value
#rets[2] = Portfolio2(dp).run_trading().portfolio_value
# rets[3] = Portfolio3(dp).run_trading().portfolio_value
# rets[4] = Portfolio4(dp).run_trading().portfolio_value
# rets[5] = Portfolio5(dp).run_trading().portfolio_value
#
# rets[6] = Portfolio6(dp).run_trading().portfolio_value
# rets[7] = Momentum(dp).run_trading().portfolio_vlaue
