1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
|
from pyisda.curve import YieldCurve, BadDay, SpreadCurve
from pyisda.utils import build_yc
from pyisda.cdsone import upfront_charge
from pyisda.legs import ContingentLeg, FeeLeg
from quantlib.settings import Settings
from quantlib.time.api import (
Date, today, Period, Years, WeekendsOnly, Quarterly, CDS, Actual360,
ModifiedFollowing, Actual365Fixed)
from quantlib.quotes import SimpleQuote
from quantlib.termstructures.credit.api import (
SpreadCdsHelper, Interpolator, ProbabilityTrait, PiecewiseDefaultCurve )
from quantlib.termstructures.yields.api import YieldTermStructure
from quantlib.pricingengines.credit.isda_cds_engine import IsdaCdsEngine
import datetime
import array
import math
import numpy as np
import pandas as pd
from db import dbengine
from yieldcurve import YC
import pdb
def get_spread(index, series):
engine = dbengine("serenitasdb")
df = pd.read_sql("SELECT date, tenor, closespread FROM index_quotes WHERE " \
"index = %(index)s and series = %(series)s",
engine,
params = {"index": index, "series": series},
parse_dates=['date'],
index_col = ['date', 'tenor'])
df = df.unstack('tenor')
df.columns = ['spread_' + c for c in df.columns.levels[1]]
return df[['spread_3yr', 'spread_5yr', 'spread_7yr', 'spread_10yr']]
def test_sc(df):
settings = Settings()
tenors = [Period(3, Years), Period(5, Years), Period(7, Years), Period(10, Years)]
quotes = []
for i in range(4):
quotes.append(SimpleQuote())
yts = YieldTermStructure()
helpers = [SpreadCdsHelper(q, t, 1, WeekendsOnly(), Quarterly,
ModifiedFollowing, CDS, Actual360(), 0.4,
discount_curve = yts,
lastperiod = Actual360(True))
for q, t in zip(quotes, tenors)]
pdc = PiecewiseDefaultCurve(ProbabilityTrait.SurvivalProbability,
Interpolator.LogLinear,
0, WeekendsOnly(), helpers, Actual365Fixed())
for r in df.itertuples(index=True, name = None):
settings.evaluation_date = Date.from_datetime(r[0])
ts = YC()
yts.link_to(ts)
for i in range(4):
quotes[i].value = r[i+1] *1e-4
return helpers
def FEP(sc, recovery):
pass
if __name__ == "__main__":
settings = Settings()
trade_date = datetime.date.today()
settings.evaluation_date = today()
start_date = datetime.date(2016, 6, 20)
term_date = datetime.date(2021, 6, 20)
dc = YC()
isda_engine = "po"
value_date = trade_date
yc = build_yc(trade_date, True)
spread = 0.007
spread = array.array('d', [spread])
value_date = datetime.date(2016, 7, 29)
step_in_date = trade_date + datetime.timedelta(days=1)
sc = SpreadCurve(trade_date, yc, start_date, step_in_date,
value_date, [term_date], spread, 0.4, True)
sc_data = sc.inspect()['data']
hazard_rate = math.log(1 + sc_data[0][1])
expiration_date = datetime.date(2016, 9, 21)
fee_leg = FeeLeg(start_date, term_date, True, 1, 1)
default_leg = ContingentLeg(start_date, term_date, 1)
fee_leg_expiry = FeeLeg(start_date, expiration_date, True, 1, 1)
default_leg_expiry = ContingentLeg(start_date, expiration_date, 1)
fee_leg_pv = fee_leg.pv(trade_date, step_in_date, trade_date,
yc, sc, True)
default_leg_pv = default_leg.pv(trade_date, step_in_date, trade_date,
yc, sc, 0.4)
V1 = default_leg_pv - fee_leg_pv
fee_leg_expiry_pv = fee_leg_expiry.pv(trade_date, step_in_date, trade_date,
yc, sc, True)
default_leg_expiry_pv = default_leg_expiry.pv(trade_date, step_in_date, trade_date,
yc, sc, 0.4)
V2 = default_leg_expiry_pv - fee_leg_expiry_pv
forward_annuity = fee_leg.pv(trade_date, expiration_date+datetime.timedelta(days=1), trade_date,
yc, sc, True)
exercise_settlement = datetime.date(2016, 9, 25)
print((0.007-0.01) * forward_annuity)
forward_price = forward_annuity*(0.007-0.01) / (
yc.discount_factor(exercise_settlement)*math.exp(-hazard_rate * (expiration_date-trade_date).days/365
))
print(fee_leg.pv(trade_date, step_in_date, trade_date,
yc, sc, True))
print(fee_leg.pv(trade_date, expiration_date + datetime.timedelta(days=1), trade_date,
yc, sc, True))
|
