practice 2

KaistLecture · KaistLecture · commit a0f40be9279a · 2016-04-20T20:59:17.000+09:00
diff --git a/new Season/Yield_Curve.txt b/new Season/Yield_Curve.txt
@@ -0,0 +1,29 @@
+2016-03-16	0.00380511
+2016-03-17	0.00380511
+2016-03-25	0.00402156
+2016-04-18	0.0044341
+2016-06-20	0.00644756
+2016-07-20	0.00635242
+2016-08-18	0.0067311
+2016-09-15	0.00725112
+2016-10-20	0.00730461
+2016-12-21	0.00789583
+2017-03-21	0.00847526
+2017-06-15	0.00896107
+2017-09-21	0.00947457
+2017-12-20	0.00992643
+2020-03-18	0.0130351
+2021-03-18	0.01416591
+2022-03-18	0.01521747
+2023-03-20	0.01613158
+2024-03-18	0.0169586
+2025-03-18	0.01767249
+2026-03-18	0.01835405
+2027-03-18	0.01898949
+2028-03-20	0.01950475
+2031-03-18	0.02076085
+2036-03-18	0.02203265
+2041-03-18	0.02262565
+2046-03-19	0.0229835
+2056-03-20	0.02298593
+2061-03-18	0.02292559
diff --git a/new Season/practice2.docx b/new Season/practice2.docx
diff --git a/new Season/swap_pricing_.py b/new Season/swap_pricing_.py
@@ -0,0 +1,126 @@
+import datetime as dt
+from dateutil.relativedelta import relativedelta
+from scipy.interpolate import interp1d
+import math
+import numpy as np
+
+#yield curve loading
+data = open("Yield_Curve.txt","r").readlines()
+data = [a.rstrip("\n").split("\t") for a in data]
+days, yields, dates = [], [], []
+for d, y in data:
+    date = dt.datetime.strptime(d,"%Y-%m-%d")
+    dates.append(date)
+    days.append((date-dates[0]).days)
+    yields.append(float(y))
+today = dates[0]
+
+#functions
+def act365(date1, date2):
+    return (date2-date1).days / 365
+def act360(date1, date2):
+    return (date2-date1).days / 360
+def spotRate(date, yieldCurve):
+    days = (date-today).days
+    return yieldCurve(days)
+def df(date, yieldCurve):
+    return math.exp(-spotRate(date,yieldCurve) 
+    * act365(today,date))        
+def fwdRate(date1, date2, daycounter, yieldCurve):
+    return 1/daycounter(date1,date2)*(df(date1,yieldCurve) / df(date2,yieldCurve) -1)
+    
+def irs_pricing(today, swaptype, notional, effectiveDate, terminationDate, 
+                fixedRate, fixedTenor, fixedDC, 
+                indexTenor, spread, floatingTenor, floatingDC, lastFixing,
+                yieldCurve):
+                    c = 1 if swaptype.upper()=="PAY" else -1
+                    #fixed leg
+                    d = effectiveDate
+                    d1 = d + fixedTenor
+                    fixedCF = []
+                    npv = 0
+                    while d1<=terminationDate:
+                        if d1>today:
+                            cf = notional * fixedDC(d,d1) * fixedRate
+                            fixedCF.append([d1, cf])
+                            npv -= cf * df(d1, yieldCurve)
+                        d = d1
+                        d1 = d + fixedTenor
+                    #floating leg
+                    d = effectiveDate
+                    d1 = d + floatingTenor
+                    floatingCF = []
+                    while d1<=terminationDate:
+                        if d1>today:
+                            rate = lastFixing if d<today else fwdRate(d,d1,floatingDC,yieldCurve)
+                            cf = notional * floatingDC(d,d1) * rate  
+                            floatingCF.append([d1, cf])
+                            npv += cf * df(d1, yieldCurve)
+                        d = d1
+                        d1 = d + floatingTenor
+                    return (c*npv, fixedCF, floatingCF)
+                    
+
+
+
+#input dates should be converted to the number of days from evaluation date
+#a fake curve
+yieldCurve = interp1d(days, yields)
+
+df_vec = np.vectorize(df)
+dfs = df_vec(dates,yieldCurve)
+#yield curve plotting
+'''
+import matplotlib.pyplot as plt
+fig, ax = plt.subplots(2,1,figsize=(8,8))
+ax[0].plot(dates,yields,"s-")
+ax[0].set_title("Yield Curve")
+ax[1].plot(dates,dfs,"d-")
+ax[1].set_title("Discount Factor")
+fig.show()
+'''
+
+
+#swap sample pricing
+notional = 10000
+swaptype = "Rec"
+today = dt.datetime(2016, 3, 16)
+effectiveDate = dt.datetime(2014, 1, 5)
+terminationDate = dt.datetime(2029, 1, 5)
+fixedRate = 0.02
+fixedDC = act365
+fixedTenor = relativedelta(months = 3)
+indexTenor = relativedelta(months = 6)
+spread = 0.001
+floatingDC = act360
+floatingTenor = relativedelta(months = 6)
+lastFixing = 0.005
+
+price = irs_pricing(today, swaptype, notional, effectiveDate, terminationDate, 
+                    fixedRate, fixedTenor, fixedDC, 
+                    indexTenor, spread, floatingTenor, floatingDC, lastFixing,
+                    yieldCurve)
+print("price = %.3f" % price[0])
+
+import matplotlib.pyplot as plt
+fig, ax = plt.subplots(2,1,figsize=(8,8))
+fixed = np.array(price[1])
+floating = np.array(price[2])
+ax[0].bar(fixed[:,0],fixed[:,1], width=60)
+ax[1].bar(floating[:,0], floating[:,1], width=60)
+fig.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
