finMath.net

# Risk Neutral Density

Getting Started

Checkout finmath-experiments from git and run maven (mvn or mvnw) from its directory. This will start a JShell. See Getting Started for details.

# Experiment 1-2: Risk neutral density

Run the following experiment from your JShell (which was launched from finmath-experiments).

## Experiment 1

The following will plot the value of a European option as a function of the strike. That is $$V(0) = E^{\mathbb{Q}}\left( V(T,K) \frac{N(0)}{N(T)}\right)$$ where $$V(T,K) = \max(S(T)-K,0)$$ is the payoff function and $$N(T) = \exp(r T)$$ is a numeraire.
JShell:

DoubleUnaryOperator value = strike -> net.finmath.functions.AnalyticFormulas.blackScholesOptionValue(100.0, 0.05, 0.20, 2.0, strike);
(new net.finmath.plots.Plot2D(0.0, 300.0, 100, value)).show();


## Experiment 2

In the following we plot the finite difference approximation of the second derivative of the value of the European option. It can be shown that the second derivative of $$V$$ with respect to $$K$$ is related to the probability density of $$S(T)$$ under $$\mathbb{Q}$$, that is $\frac{\partial V(T,K)}{\partial K} \ = \ \phi_{S(T)}(K) \ \frac{N(0)}{N(T)} \text{.}$

JShell:

import net.finmath.functions.AnalyticFormulas;
import net.finmath.plots.Plot2D;

DoubleUnaryOperator value = strike -> AnalyticFormulas.blackScholesOptionValue(100.0, 0.05, 0.20, 2.0, strike);

double h = 1E-2;
DoubleUnaryOperator secondDerivative = strike -> (value.applyAsDouble(strike+h) - 2*value.applyAsDouble(strike) + value.applyAsDouble(strike-h))/(h*h);

Plot2D plot = new Plot2D(0.0, 300.0, 100, secondDerivative);
plot.setYAxisNumberFormat(new java.text.DecimalFormat("0.0E00"));
plot.setXAxisLabel("strike").setYAxisLabel("frequency");
plot.show();


## Summary (Experiments 1-2)

We have created a plot of the value and the second derivative of the Black-Scholes formula. The second derivative of the value with respect to the price is (up to a scaling) the probability density of the underlying stock at maturity (under the terminal measure).

## Where to continue

You may continue with