[Quantlib-users] Caibrating HW 1F model to Swaption vol

[Ashwani S. <ash...@gm...>]

Looked into archeived messages but couldn't find anything relevant. I am
trying to calibrate Hull White model using Goutam's code here
<http://gouthamanbalaraman.com/blog/short-interest-rate-model-calibration-quantlib.html>
with
a tweak that instead of using flatForward term_structure, I am using
ql.PiecewiseCubicZero(calc_date, bond_helpers, day_count) calibrated to US
Treasuries.

I suspect the error has to do with negative rates, but here is the piece of
code which throws an error when I try running it

    swaptions = []
    index = ql.USDLibor(ql.Period(3, ql.Months), term_structure)

    # Read US Swaptions vol data
    CalibrationData = namedtuple("CalibrationData",
                             "start, length, volatility")
    data = [CalibrationData(1, 5, 0.1148),
            CalibrationData(2, 4, 0.1108),
            CalibrationData(3, 3, 0.1070),
            CalibrationData(4, 2, 0.1021),
            CalibrationData(5, 1, 0.1000 )]

    for d in data:
        vol_handle = ql.QuoteHandle(ql.SimpleQuote(d.volatility))
        helper = ql.SwaptionHelper(ql.Period(d.start, ql.Years),
                                   ql.Period(d.length, ql.Years),
                                   vol_handle,
                                   index,
                                   index.tenor(),
                                   day_count,
                                   day_count,
                                   term_structure,
                                   ql.Normal,
                                   )
        helper.setPricingEngine(engine)
        swaptions.append(helper)

    optimization_method = ql.LevenbergMarquardt(1.0e-8,1.0e-8,1.0e-8)
    end_criteria = ql.EndCriteria(10000, 100, 1e-6, 1e-8, 1e-8)
    model.calibrate(swaptions, optimization_method, end_criteria)


---------------------------------------------------------------------------RuntimeError
                             Traceback (most recent call
last)<ipython-input-78-1c3c09dff89b> in <module>     24 model =
ql.HullWhite(yield_curve_handle)     25 engine =
ql.JamshidianSwaptionEngine(model)---> 26 alpha, sigma =
calibrate_hullwhite(yield_curve_handle, engine, model, day_count)
27      28 # Generate HW simulation paths
<ipython-input-77-cab060688d5e> in calibrate_hullwhite(term_structure,
engine, model, day_count)     32     optimization_method =
ql.LevenbergMarquardt(1.0e-8,1.0e-8,1.0e-8)     33     end_criteria =
ql.EndCriteria(10000, 100, 1e-6, 1e-8, 1e-8)---> 34
model.calibrate(swaptions, optimization_method, end_criteria)     35
   36     return model.params()
~\AppData\Local\anaconda3\lib\site-packages\quantlib-1.21-py3.8-win-amd64.egg\QuantLib\QuantLib.py
in calibrate(self, *args)  11289 > 11291         return
_QuantLib.CalibratedModel_calibrate(self, *args)

  11290     def calibrate(self, *args):

11292 11293 def setParams(self, params): RuntimeError: strike +
displacement (-0.0114291 + 0) must be non-negative



Any suggestions would be appreciated.

Thanks.