Quant Python ToolKit

This package is intended to be a layer above QuantLib Python and a few other quantitative libraries to be more accessible for quantitative finance calculations.

Minimal Example

Here is a minimal example for valuing a bond using a provided zero rates.

from qtk import Controller, Field as F, Template as T

data = [{
          'Compounding': 'Compounded',
          'CompoundingFrequency': 'Annual',
          'Currency': 'USD',
          'DiscountBasis': '30/360',
          'DiscountCalendar': 'UnitedStates',
          'ListOfDate': ['1/15/2015', '7/15/2015', '1/15/2016'],
          'ListOfZeroRate': [0.0, 0.005, 0.007],
          'ObjectId': 'USD.Zero.Curve',
          'Template': 'TermStructure.Yield.ZeroCurve'},
         {
          'DiscountCurve': '->USD.Zero.Curve',
          'ObjectId': 'BondEngine',
          'Template': 'Engine.Bond.Discounting'},
         {
          'AccrualCalendar': 'UnitedStates',
          'AccrualDayConvention': 'Unadjusted',
          'AsOfDate': '2016-01-15',
          'Coupon': 0.06,
          'CouponFrequency': 'Semiannual',
          'Currency': 'USD',
          'DateGeneration': 'Backward',
          'EndOfMonth': False,
          'IssueDate': '2015-01-15',
          'MaturityDate': '2016-01-15',
          'ObjectId': 'USD.TBond',
          'PaymentBasis': '30/360',
          'PricingEngine': '->BondEngine',
          'Template': 'Instrument.Bond.TreasuryBond'}]

res = Controller(data)
asof_date = "1/15/2015"

ret = res.process(asof_date)
tbond = res.object("USD.TBond")
print tbond.NPV()

The basic idea here is that once you have the data prepared, the Controller can be invoked to do the calculations. A few points that are worth noting here.

• All the data is textual and rather intuitive. For instance, the coupon frequency is just stated as Annual or Semiannual. The same is true for a lot of other fields. For dates, the dateutil package is used to parse and covers a wide variety of formats.

• The data is essentially a list of dict with each dict corresponding to a specific object as determined by the value to the key Template in each dict. Each object here has a name as specified by the value of the key ObjectId

• One of the values can refer to another object described by a dict by using the reference syntax. For instance, the first dict in the data list (with ObjectId given as USD.Zero.Curve ) variable refers to an interest rate term structure of zero rates. The next object is a discounting bond engine, and require an yield curve as input for the discount curve. Here the yield curve is refered by using the prefix -> along with the name of the object we are referring to.

• Here, the Controller parses the data, and figures out the dependency and processes the object in the correct order and fulfills the dependencies behind the scenes.

Introspection

There are a few convenience methods that provide help on how to construct the data packet. For example, the help method in the template prints out the summary and list of fields on how to construct the data packet for the template.

> T.TS_YIELD_BOND.help()

Description

A template for creating yield curve by stripping bond quotes.

Required Fields

• Template [Template]: 'TermStructure.Yield.BondCurve'
• InstrumentCollection [List]: Collection of instruments
• AsOfDate [Date]: Reference date or as of date
• Country [String]: Country
• Currency [String]: Currency

Optional Fields

• ObjectId [String]: A unique name or identifier to refer to this dictionary data
• InterpolationMethod [String]: The interpolation method can be one of the following choices: LinearZero, CubicZero, FlatForward, LinearForward,LogCubicDiscount.
• DiscountBasis [DayCount]: Discount Basis
• SettlementDays [Integer]: Settlement days
• DiscountCalendar [Calendar]: Discount Calendar

The help method prints the description in info method in Markdown format. While using IPython/Jupyter notebooks, the description prints in a nice looking format. One can start with a sample data packet to fill out the input fields using the sample_data method.

> T.TS_YIELD_BOND.sample_data()

{'AsOfDate': 'Required (Date)',
 'Country': 'Required (String)',
 'Currency': 'Required (String)',
 'DiscountBasis': 'Optional (DayCount)',
 'DiscountCalendar': 'Optional (Calendar)',
 'InstrumentCollection': 'Required (List)',
 'InterpolationMethod': 'Optional (String)',
 'ObjectId': 'Optional (String)',
 'SettlementDays': 'Optional (Integer)',
 'Template': 'TermStructure.Yield.BondCurve'}

Installation

You can install qtk using pip or easy_install

pip install qtk

or

easy_install qtk

qtk has a dependency on QuantLib-Python which needs to be installed as well.