FxPaul

Math in finance or vice versa

Correlation trading on FX

leave a comment »

Correlation trading is based on few simple ideas:

  1. Correlations are changing with time
  2. Correlations of pairs of 3 currencies are bounded by strict equation
  3. Correlation is bound in the interval [-1, 1]

Therefore, one may try to “buy correlation” at -1 and to “sell it” at +1.

Read the rest of this entry »

Written by fxpaul

November 1, 2011 at 13:04

Exponential Ornstein-Uhlenbeck process and USD/CHF

leave a comment »

In the previous post USD/CHF is considered to follow a mean-reverting process. Let’s look at the dynamics of the process during the days of year 2010.

Calibration procedure

Calibration is simple: get logs of prices and calibrate against classic Ornstein-Uhlenbeck process as it is described here.
Read the rest of this entry »

Written by fxpaul

July 25, 2011 at 14:19

Posted in trading math

Tagged with

SABR model calibration

leave a comment »

Don’t use this!

The SABR model is a stochastic volatility model, which attempts to capture the volatility smile in derivatives markets. The name stands for “Stochastic Alpha, Beta, Rho”, referring to the parameters of the model. It was developed by Patrick Hagan, Deep Kumar, Andrew Lesniewski, and Diana Woodward.

The SABR model describes a single forward F, such as a LIBOR forward rate, a forward swap rate, or a forward stock price. The volatility of the forward F is described by a parameter σ. SABR is a dynamic model in which both F and σ are represented by stochastic state variables whose time evolution is given by the following system of stochastic differential equations:
dF_t = \sigma_t F_t^{\beta} dW_t
d\sigma_t = \alpha\sigma_t dZ_t
Constant parameters should satisfy the condition 0 \leq \beta \leq 1, \alpha \geq 0
Here, W_t and Z_t are two correlated Wiener processes with correlation coefficient -1\leq\rho\leq 1. For simplicity sake, we assume that \rho = 1, therefore, we put Z_t = W_t:
dF_t = \sigma_t F_t^{\beta} dW_t
d\sigma_t = \alpha\sigma_t dW_t

Read the rest of this entry »

Written by fxpaul

June 17, 2011 at 09:27

Posted in trading math

Is EUR/USD mean reverting?

with 2 comments

Nassim Taleb in his book “Dynamic Hedging” provides an empirical rule to validate if the market has mean reversion tendency. It states that if volatility lowers if the longer time frame is used, then the mean reversion takes place.
Read the rest of this entry »

Written by fxpaul

June 16, 2011 at 20:52

Posted in trading math

Exponential Ornstein-Uhlenbeck process

with 3 comments

Let’s consider parameter estimation for the following modification of Ornstein-Uhlenbeck process:
dS_t = \theta(\mu - S_t) dt + \sigma dW_t
P_t= \exp S_t
This model is simplification of Schwarz Model 1, one of Short-rate models
Read the rest of this entry »

Written by fxpaul

June 8, 2011 at 08:28

Posted in trading math

Closed-form solution of modified Ornstein-Uhlenbeck process

with 11 comments

Process definition

In this article we deduce the closed-from solution of the modified version of Ornstein-Uhlenbeck process:
dS = \theta(\mu - S) dt + \sigma S dW_t
where \theta – mean reversion parameter, \mu – mean and \sigma – volatility.

Integrating factor approach

There exists a general approach to non-linear stochastic differential equations of the form:
dX_t = f(t,X_t) dt + c(t) X_t dW_t \, , \, X_0 = x
where f and g are given continuous and deterministic functions.

The method consists of:

  1. Define the integrating factor:
    F_t = F_t(\omega) = \exp\left( - \int_0^t {c(s) dW_s} + \frac{1}{2}\int_0^t {c^2(s) dW_s} \right)
  2. So the original equation could be written as d\left(F_t X_t\right) = F_t f(t,X_t) dt
  3. Now define
    Y_t(\omega) = F_t(\omega) X_t(\omega)
    so that X_t = F^{-1}_t Y_t
  4. And it yields the deterministic differential equation for each \omega\in\Omega
    \frac{dY_t(\omega)}{dt} = F_t(\omega) f\left(t, F^{-1}_t(\omega) Y_t(\omega)\right); \, Y_0 = x

We can therefore solve it with \omega as a parameter to find Y_t(\omega) and then obtain X_t(\omega)

Read the rest of this entry »

Written by fxpaul

May 27, 2011 at 08:39

Posted in trading math

Calibration of Ornstein-Uhlenbeck process

with one comment

Introduction

In mathematics, Ornstein-Uhlenbeck process satisfies the following stochastic differential equation:
dS = \theta(\mu - S) dt + \sigma dW_t
where \theta – mean reversion parameter, \mu – mean and \sigma – volatility.

In finance, it is used to model interest rates, currency exchange rates and commodity prices. Although it is usually modified to incorporate non-negativity of prices.

Ordinary Least-Squares Approach to calibration

The simplest approach to the calibration problem is to convert SDE to finite difference equation (as it is usually used in Monte Carlo simulation) and to rearrange parts to Ordinary Least Squares equation.

The simplest updating formula for Ornstein-Uhlenbeck process is:
S_t - S_{t-1} = \theta (\mu - S_{t-1}) \Delta t + \sigma \sqrt{\Delta t} W_{t-1}
By rearranging we obtain:
S_t - S_{t-1} = \theta\mu\Delta t - \theta S_{t-1}\Delta t +  \sigma \sqrt{\Delta t} W_{t-1}
Comparing with simple regression formula:
y = a + bx + \epsilon
we can equate as follows:
y = S_t - S_{t-1}
x = S_{t-1}
a = \theta\mu\Delta t
b = -\theta\Delta t
\epsilon = \sigma \sqrt{\Delta t} W_{t-1}
and immediately obtain the following:
\theta = -\frac{b}{\Delta t}
\mu    = -\frac{a}{b}

As W_t is drawn from normal distribution, its expectation equals zero and one should use variance to obtain \sigma:
\sigma_{\epsilon} = \sigma \sqrt{\Delta t} \sigma_W
where \sigma_{W} = 1 as it has been already normalized by \sqrt{\Delta t}. Finally, we can obtain:
\sigma = \frac{\sigma_{\epsilon}}{\sqrt{\Delta t}}

So, regression of S_{t-1} against S_t - S_{t-1} gives estimation of process parameters.

The modified process

Let’s consider the process with slight modification and apply the same approach to the modified process:
dS = \theta(\mu - S) dt + \sigma S dW_t
Then the naive updating formula is
S_t - S_{t-1} = \theta (\mu - S_{t-1}) \Delta t + \sigma S_{t-1} \sqrt{\Delta t} W_{t-1}
Then dividing by S_{t-1}:
\frac{S_t}{S_{t-1}} - 1 = \theta\mu\Delta t\frac{1}{S_{t-1}} - \theta \Delta t +  \sigma \sqrt{\Delta t} W_{t-1}
Given simple regression formula:
y = a + bx + \epsilon
we can equate as follows:
y = \frac{S_t}{S_{t-1}} -1
x = \frac{1}{S_{t-1}}
a = - \theta\Delta t
b = \theta\mu\Delta t
\epsilon = \sigma \sqrt{\Delta t} W_{t-1}
and immediately obtain the following:
\theta = - \frac{a}{\Delta t}
\mu    = - \frac{b}{a}
Applying the same logic as in previous section, finally we get:
\sigma = \frac{\sigma_{\epsilon}}{\sqrt{\Delta t}}

Therefore, regression of \frac{S_t}{S_{t-1}} - 1 against \frac{1}{S_{t-1}} yields estimation of modified process parameters.

Open questions

  1. Bias of the estimators. For the original process this approach usually gives quite precise estimation of mean and volatility but fails to provide mean reversion parameter
  2. Closed-form solution of the modified SDE. It could be used to improve the updating formula
  3. Statistical hypothesis testing if the sample drawn from the process. This is quite crucial point as it helps to identify model regime shift in trading.

Written by fxpaul

May 26, 2011 at 11:17

Posted in trading math

Quantum Mechanics and Market Observables

leave a comment »

Quantum mechanics has quite powerful notion of observables. They are just properties of system state that can be determined by measurement process. Let’s recall main principles of quantum mechanics and how they could be applied to markets.

Read the rest of this entry »

Written by fxpaul

January 15, 2011 at 10:53

Common trading task – Part 1

leave a comment »

Stochastic process

As Wikipedia suggests, a stochastic process is a random process, the counterpart to a deterministic process. For our simple tasks all we need is a time series, i.e. for each moment of time we have only one random value, or price: P(t). The process has a definite starting point P(0) but its further evolution has some degree of uncertainty described by probability distribution.

A lot of types of stochastic processes has been studied in mathematical literature. In this article I use only Itō processes as they provide quite good approximation of price dynamics.

Read the rest of this entry »

Written by fxpaul

January 3, 2011 at 12:00

Posted in trading math

Tagged with ,

Mixing up Fx and Stocks

leave a comment »

Anybody can find enormous amount of books about stocks trading but all of Forex books give an expression of spurious market: plain old strategies, positive only examples of profit-loss calculations etc. It’s easy to get impressed as well as to get depressed by results of trading.

I want to explore how we could apply stock strategies to foreign exchange markets. Vast majority of stock strategies should be adopted to turbulent dynamic of Forex markets.

The nearest future is:

  1. Define the common trading task in a precise, mathematical way – this is the starting point.
  2. Monte Carlo methods for optimization and trading experiments.
  3. Simple strategy emulations on model processes

I do hope the list will be continued.

Written by fxpaul

December 27, 2010 at 20:56

Posted in Uncategorized

Tagged with ,