# -*- coding: utf-8 -*-

"""

Created on Wed Aug 22 15:14:47 2018

Part 1 creating the multi-flavored option backtests: SAN

@author: dpsugasa

"""

import os, sys

import pandas as pd

from tia.bbg import LocalTerminal

import numpy as np

from datetime import datetime

from operator import itemgetter

import plotly

import plotly.plotly as py #for plotting

import plotly.graph_objs as go

import plotly.dashboard_objs as dashboard

import plotly.tools as tls

import plotly.figure_factory as ff

import credentials

#set the script start time

start_time = datetime.now()

#plotly function

def create_trace(df, label):

return trace

# set dates to grab data

start_date = '01/01/2012'

end_date = "{:%m/%d/%Y}".format(datetime.now())

date_now = "{:%m_%d_%Y}".format(datetime.now())

# Files in Option Directory

path2 = "C:\\Users\\dpsugasa\\WorkFiles\\option_backtests\\options"

dirs2 = os.listdir(path2)

#path = f"D:\\Users\\dpsugasa\\option_backtests\\options\\{direct}"

#dirs = os.listdir(path)

ts = {} #dict for trades

tkr = {} #dict for tickers

px = {} #dict of original dataframes per ID

px2 = {} #final dict of a prices

temp = {} #dict of temp dataframes

temp2 = {} #additional dict of temp dataframes

tot_ser = {} #df for all options series per strategy

m = {} #df for option ref data

n = {} #temp df

quants= []

for direct in dirs2:

path = f"C:\\Users\\dpsugasa\\WorkFiles\\option_backtests\\options\\{direct}"

dirs = os.listdir(path)

for file in dirs:

curr_path = fr'C:\Users\dpsugasa\Workfiles\option_backtests\options\{direct}/{file}'

ts[file] = pd.read_csv(curr_path)

ts[file].Date = pd.to_datetime(ts[file].Date)

ts[file].Expiry = pd.to_datetime(ts[file].Expiry)

ts[file] = ts[file].drop('Vol', axis=1)

ts[file]['Amount'] = ts[file]['Amount']/100

tkr[file] = ts[file]['Ticker'].unique().tolist()

# set securities, and fields

IDs = tkr[file]

fields = ['LAST PRICE']

d = {} #dict of original dataframes per ID

d2 = {} #final dict of a prices

temp = {} #dict of temp dataframes

temp2 = {} #additional dict of temp dataframes

ref_data = ['OPTION_ROOT_TICKER', 'OPT_MULTIPLIER', 'OPT_UNDL_PX',

'COMPANY_CORP_TICKER', 'CRNCY']

#get initial prices in 'd', create a temp dataframe with entry/exit dates,

# price, and expiry for each ticker

for name in IDs:

d[file, name] = LocalTerminal.get_historical(name, fields, start_date, end_date,

period = 'DAILY').as_frame()

d[file, name].columns = d[file, name].columns.droplevel()

d[file, name] = d[file, name].fillna(method = 'ffill')

temp[file, name] = ts[file].loc[ts[file].Ticker == name][['Date',

'Amount', 'Expiry', 'Direction','Shares']]

temp[file, name].index = temp[file, name].Date

temp[file, name] = temp[file, name].drop('Date', axis=1)

m[file, name] = LocalTerminal.get_reference_data(name, ref_data).as_frame()

n[file] = LocalTerminal.get_reference_data(name, ref_data).as_frame()

#set option qtty equal to $1mm USD worth of bonds so they can be compared in 'return space'

opt_curr = n[file]['CRNCY'].item() + " CURNCY"

curr_px = LocalTerminal.get_reference_data(opt_curr, 'PX_LAST').as_frame().values.item()

multy = 100.00 #n[file]['OPT_MULTIPLIER'].item() Hard coding as 100 multiplier

undl = n[file]['OPT_UNDL_PX'].item()

bond_size = 1000000.0 #1m USD

b_size_adj = bond_size/curr_px

opt1_qtty = np.round(((b_size_adj)/(multy*undl)))

for l in IDs:

quants.append(opt1_qtty)

#because some of the price info does not extend to maturity, make new pricing

# dataframes that have the full price set, including expiry value = 'd2'

for i in IDs:

temp2[file, i] = pd.DataFrame(np.nan, columns = ['LAST PRICE_NA'],

index = pd.date_range(start = d[file, i].index[0],

end = temp[file,i]['Expiry'][-1],

freq = 'B'))

frames = [temp2[file, i], d[file, i]]

d2[file, i] = pd.concat(frames, join = 'outer', axis = 1)

d2[file, i] = d2[file, i].drop(['LAST PRICE_NA'], axis = 1)

#making sure entry and exit days are in the index

d2[file, i].loc[temp[file, i].index[-1]] = np.nan

d2[file, i].loc[temp[file, i].index[0]] = np.nan

d2[file, i] = d2[file, i].sort_index()

d2[file, i] = d2[file, i].fillna(method = 'ffill')

d2[file, i] = d2[file, i].dropna()

d2[file, i] = d2[file, i].sort_index().truncate(after = end_date)

d2[file, i]['trade'] = 0.0

d2[file, i]['prev_pos'] = 0.0

d2[file, i]['pos'] = 0.0

#entry trade; 1.0 for buy, -1.0 for sell; amend price to entry price

if temp[file, i]['Direction'][0] == 'Buy':

d2[file, i]['trade'].loc[temp[file, i].index[0]] = 1.0

#d2[i]['LAST PRICE'].loc[temp[i].index[0]] = temp[i]['Amount'].loc[temp[i].index[0]]

else:

d2[file, i]['trade'].loc[temp[file, i].index[0]] = -1.0

#d2[i]['LAST PRICE'].loc[temp[i].index[0]] = temp[i]['Amount'].loc[temp[i].index[0]]

#exit trade; current options use the final day of the series

if temp[file, i]['Expiry'][-1] < pd.to_datetime(end_date):

if temp[file, i]['Direction'][-1] == 'Buy':

d2[file, i]['trade'].loc[temp[file, i].index[-1]] = 1.0

d2[file, i]['LAST PRICE'].loc[temp[file, i].index[-1]] =\

temp[file, i]['Amount'].loc[temp[file, i].index[-1]]

else:

d2[file, i]['trade'].loc[temp[file, i].index[-1]] = -1.0

d2[file, i]['LAST PRICE'].loc[temp[file, i].index[-1]] =\

temp[file, i]['Amount'].loc[temp[file, i].index[-1]]

else:

if temp[file, i]['Direction'][-1] == 'Buy':

d2[file, i]['trade'][-1] = 1.0

else:

d2[file, i]['trade'][-1] = -1.0

d2[file, i] = d2[file, i].sort_index()

for z, row in d2[file, i].iterrows():

idx_loc = d2[file, i].index.get_loc(z)

prev = idx_loc -1

d2[file, i]['prev_pos'].loc[z] = d2[file, i]['pos'].iloc[prev]

if row['trade'] == 1.0:

d2[file, i]['pos'].loc[z] = 1.0 + d2[file, i]['prev_pos'].loc[z]

elif row['trade'] == -1.0:

d2[file, i]['pos'].loc[z] = -1.0 + d2[file, i]['prev_pos'].loc[z]

else:

d2[file, i]['pos'].loc[z] = d2[file, i]['prev_pos'].loc[z]

d2[file, i]['shares'] = temp[file, i]['Shares'].iloc[0]

d2[file, i]['qtty'] = opt1_qtty

d2[file, i]['cash_val'] = 0.0

d2[file, i]['trade_val'] = d2[file, i]['trade']*d2[file, i]['shares']*\

d2[file, i]['qtty']*d2[file, i]['LAST PRICE']

d2[file, i]['pos_val'] = d2[file, i]['prev_pos']*d2[file, i]['shares']*\

d2[file, i]['qtty']*d2[file, i]['LAST PRICE']

for z, row in d2[file, i].iterrows():

idx_loc = d2[file, i].index.get_loc(z)

prev = idx_loc -1

if row['trade'] != 0:

d2[file, i]['cash_val'].loc[z] =\

np.negative(d2[file, i]['trade'].loc[z]* \

d2[file, i]['shares'].loc[z]* \

d2[file, i]['qtty'].loc[z]* \

d2[file, i]['LAST PRICE'].loc[z]) +\

d2[file, i]['cash_val'].iloc[prev]

else:

d2[file, i]['cash_val'].loc[z] = d2[file, i]['cash_val'].iloc[prev]

d2[file, i]['total_val'] = d2[file, i]['trade_val'] +\

d2[file, i]['pos_val'] + d2[file, i]['cash_val']

d2[file, i] = d2[file, i].truncate(after = end_date)

frames = []

for i in IDs:

frames.append(d2[file, i]['total_val'])

tot_ser[file] = pd.concat(frames, join='outer', axis=1)

tot_ser[file] = tot_ser[file].fillna(method = 'ffill')

tot_ser[file] = tot_ser[file].fillna(0)

tot_ser[file]['port_val'] = tot_ser[file].sum(axis=1)/b_size_adj

# tot_ser[file]['port_val'] = tot_ser[file].sum(axis=1)

root_tkr = m[file, i]['COMPANY_CORP_TICKER'].item()

file_name = file.replace('.csv','')

trace1 = go.Scatter(

x = tot_ser[file]['port_val'].index,

y = tot_ser[file]['port_val'].values,

name = f'{root_tkr}_Options',

line = dict(

color = ('#4155f4'),

width = 2))

layout = {'title' : f'{file_name}_{date_now}',

'xaxis' : {'title' : 'Date', 'type': 'date'},

'yaxis' : {'title' : 'Returns'},

# 'shapes': [{'type': 'rect',

# 'x0': d[i]['scr_1y'].index[0],

# 'y0': -2,

# 'x1': d[i]['scr_1y'].index[-1],

# 'y1': 2,

# 'name': 'Z-range',

# 'line': {

# 'color': '#f48641',

# 'width': 2,},

# 'fillcolor': '#f4ad42',

# 'opacity': 0.25,

# },]

}

data = [trace1]

figure = go.Figure(data=data, layout=layout)

py.iplot(figure, filename =\

f'option_backtest/{root_tkr}/{file_name}/{file_name}_{date_now}')

b = {}

for file in dirs:

b[file] = create_trace(tot_ser[file]['port_val'], file)

root_tkr = n[file]['COMPANY_CORP_TICKER'].item()

file_name = file.replace('.csv','')

layout = {'title' : f'{root_tkr}_All_Flavors',

'xaxis' : {'title' : 'Date', 'type': 'date'},

'yaxis' : {'title' : 'Returns'},

# 'shapes': [{'type': 'rect',

# 'x0': d[i]['scr_1y'].index[0],

# 'y0': -2,

# 'x1': d[i]['scr_1y'].index[-1],

# 'y1': 2,

# 'name': 'Z-range',

# 'line': {

# 'color': '#f48641',

# 'width': 2,},

# 'fillcolor': '#f4ad42',

# 'opacity': 0.25,

# },]

}

data = list(b.values())

figure = go.Figure(data=data, layout=layout)

py.iplot(figure, filename =\

f'option_backtest/{root_tkr}/{root_tkr}_All_Flavors_{date_now}')

print ("Time to complete:", datetime.now() - start_time)