Pyinsurance 1.1.9

Tools to backtest and create your own portfolio insurance strategy

Create your own insured portfolio using several Tools.

First ,To install, just use pip :

pip install pyinsurance

Required Dependencies are listed below , such :

Dependency	Version
arch	5.0.1
numpy	1.20.1
scipy	1.6.2
statsmodels	0.12.2
numba	0.52.1
setuptools	60.5.0
pandas	1.2.4
pyvar	0.0.1

There is no dependency verification , so please, make sure to have installed every required one before using the package.

Example

To begin, let’s extract some included default data :

import pyinsurance
from pyinsurance.pymolder import tipp_model
from pyinsurance.data.IRX import load as d1
from pyinsurance.data.sp500 import load as d2
import matplotlib.pyplot as plt

risky_Asset = d2()
safe_Asset = d1()/52  #we divided by 52 as we use weekly rates

Let’s initalise our first insured portfolio now!

For instance,we set our lock-in rate , minimum capital risk allocation , threshold for capital injection , allocate funds ,strategy’s percentage floor ,multipler,benchmark returns and rebalancement cycle being respectively equal to :

lock_in_rate = 0.05
mcr = 0.40
tfci = 0.80
fund = 100
floor = 0.80
multiplier = 10
Benchmark_return = risk_Asset
Rebalancement_frequency = 52 # once a week -> 52 weeks a year

Running the tipp_model class :

res = tipp_model(risk_Asset,safe_Asset,lock_in_rate,mcr,tfci,fund,\
                floor,multiplier,risk_Asset,Rebalancement_frequency)

Our strategy-insured backtest is ready !

import matplotlib.pyplot as plt
from pyinsurance.Metric_Generator.returns_metrics import Cumulative_ret


fig = plt.figure(figsize=(15,5))
ax0 = fig.add_subplot(111)
plt.plot(risk_Asset.index,Cumulative_ret(risk_Asset)*100,label = 'Non-Insured Performance')
plt.plot(risk_Asset.index,res.Fund,label = 'Fund Performance')
plt.plot(risk_Asset.index,res.Reference_capital,label = 'Reference Capital',linestyle="--")
plt.plot(risk_Asset.index,res.floor,label = 'Floor',linestyle="-.")
plt.legend()
plt.show()

And our capital injections through the period are presented as:

fig = plt.figure(figsize=(15,5))
ax1 = fig.add_subplot(111)
plt.plot(risk_Asset.index,res.capital_reinjection,label = 'Injected Capital')
plt.legend()
plt.show()

If you want to backtest the VaR, you can use the varpy library:

import pyvar
from varpy.Backtester.bktst import Backtest
from varpy.Backtester.time_Significance import Testing
VaR , CVaR = Backtest(data, 500, 2, 0.05, model = 'EVT')

fig = plt.figure(figsize=(15,5))
plt.plot(data[500:])
plt.plot(VaR, label = 'VaR')
plt.plot(CVaR, label = 'CVaR')
plt.legend()
plt.show()