def get_prices(ticker, start_date):
prices_df = norgatedata.price_timeseries(ticker,
start_date=start_date,
format=NORGATE_TIMESERIESFORMAT)
prices_df['TICKER'] = ticker
prices_df.index.rename('DATE', inplace=True)
prices_df = drop_columns_from_norgate_dataset(prices_df)
return prices_df.groupby(['TICKER', 'DATE']).min()
def load_ng_historical(symbol,startdate=str_to_dt('1970-01-01'),enddate=None,interval="D"):
pricedata = ng.price_timeseries(
symbol,
stock_price_adjustment_setting = ng.StockPriceAdjustmentType.TOTALRETURN,
padding_setting=ng.PaddingType.ALLWEEKDAYS,
start_date = startdate,
end_date = enddate,
interval=interval,
format='pandas-dataframe'
)
return pricedata
def hist_sym(sym, start):
#sym: symbol or ticker
#start: start date 'yyyy-mm-dd' format
priceadjust = norgatedata.StockPriceAdjustmentType.TOTALRETURN
padding_setting = norgatedata.PaddingType.NONE
timeseriesformat = 'pandas-dataframe'
start_date = pd.Timestamp(start)
try:
dff = norgatedata.price_timeseries(
sym,
stock_price_adjustment_setting=priceadjust,
padding_setting=padding_setting,
start_date=start_date,
format=timeseriesformat)
return (dff)
except:
return ('no symbol')
def norgate_defined_start(watchlist, start_date, end_date, frequency):
symbols = norgatedata.watchlist_symbols(watchlist)
df_list = []
for symbol in symbols:
norgate_df = norgatedata.price_timeseries(symbol,
start_date=start_date,
end_date=end_date,
interval=frequency,
format='pandas-dataframe')
# creating a symbol column so that we can identify
# what ticker information we're looking at
norgate_df['Symbol'] = symbol
# appending the dataframes to an empty list to concatenate them after the for loop
df_list.append(norgate_df)
return df_list
def create_index(start, end, index_ticker):
symbol = index_ticker
norgate_df = norgatedata.price_timeseries(symbol,
start_date=start,
end_date=end,
interval='D',
format='pandas-dataframe')
norgate_df.drop([
'Open', 'High', 'Low', 'Volume', 'Turnover', 'Unadjusted Close',
'Dividend'
],
axis=1,
inplace=True)
# changing name from Close to whatever the ticker is
norgate_df.rename({'Close': index_ticker}, inplace=True, axis=1)
# return the cumulative pct return over the period we have the data for
norgate_df[index_ticker] = np.cumprod(
1 + norgate_df[index_ticker].pct_change())
return norgate_df
from datetime import datetime
lookback_period = 60
end_date = datetime.now()
watchlistname = 'spy_tlt_gld'
symbols = norgatedata.watchlist_symbols(watchlistname)
df_list = []
for symbol in symbols:
norgate_df = norgatedata.price_timeseries(
symbol,
end_date=end_date,
limit=lookback_period,
interval='D',
format='pandas-dataframe')
# creating a symbol column so that we can identify
# what ticker information we're looking at
norgate_df['Symbol'] = symbol
# appending the dataframes to an empty list to concatenate them after the for loop
df_list.append(norgate_df)
# joins the dataframes together for all tickers
appended_data = pd.concat(df_list)
# just keep the closing data and the symbol