Quantlib Yield Curve

Is it possible to create yield curve object in Quantlib given some function of time? For example, given Nelson-Siegel parameters, create yield curve which can compute zero yield for any date >= reference date.

Yes, it is possible. You can feed bond instruments with prices and fit Nelson-Siegel parameters.

Try this simple example:

import QuantLib as ql
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

today = ql.Date(30,12,2019)
ql.Settings.instance().evaluationDate = today

bonds = pd.DataFrame({
'maturity': ['30-03-2020','30-12-2020', '15-06-2021', '17-10-2022', '17-04-2023'],
'coupon': [0,1.25, 3.40, 6.40, 2.25],
'price': [99.88, 99.75, 101.98, 108.70, 96.46]
})

calendar = ql.TARGET()
bondSettlementDays = 2
frequency = ql.Annual
dc = ql.ActualActual(ql.ActualActual.ISMA)
accrualConvention = ql.ModifiedFollowing
convention = ql.ModifiedFollowing
redemption = 100.0

instruments = []
for idx, row in bonds.iterrows():
schedule = ql.Schedule(
bondSettlementDate,
ql.Date(row.maturity, '%d-%m-%Y'),
ql.Period(frequency),
calendar,
accrualConvention,
accrualConvention,
ql.DateGeneration.Backward,
False)
helper = ql.FixedRateBondHelper(
ql.QuoteHandle(ql.SimpleQuote(row.price)),
bondSettlementDays,
100.0,
schedule,
[row.coupon / 100],
dc,
convention,
redemption)

instruments.append(helper)

curveSettlementDays = 2
curveFittingMethod = ql.NelsonSiegelFitting()
tolerance = 1.0e-5
iterations = 1000

curve_ns = ql.FittedBondDiscountCurve(curveSettlementDays, calendar, instruments,
dc, curveFittingMethod, tolerance, iterations)
t = np.linspace(1,3.25)
rates_ns = [curve_ns.zeroRate(n, ql.Annual).rate() for n in t]
plt.figure(figsize=(15,5)),
plt.plot(t, rates_ns);