3
$\begingroup$

I have a set of Langevin equations given by

$${\mathbf{\dot{x}}} = \mathbf{-Q \,x} + \mathbf{\eta} \tag{1}$$

where $\eta$ is white Gaussian noise and $Q$ is not a function of $x$.

Using Euler's method for SDE, I generated a time series of $\bf{x}$, which (as expected) is noisy due to $\eta$. Using the time series of $\bf{x}$, generating $\langle(\bf{-Qx} + \bf{\eta})(\bf{-Qx} + \bf{\eta)^T}\rangle$ gave me a finite matrix.

However, for the case where I don't know $\bf{Q}$ and $\eta$, I need to recover $\bf\dot{x}$ from the noisy time series $\bf x$. How do I get $\langle\bf\dot{x}\dot{x}^T\rangle$ from the generated $\bf x$? I hope this can be done and would give me the same value for $\langle(\bf{-Qx} + \bf{\eta})(\bf{-Qx} + \bf{\eta)^T}\rangle$.

Equation 1 has the solution

$${x} (t) = e^{-Qt}x(0) +\int_{0}^{t}dt'e^{-(t-t)'Q}\eta(t')$$

but now I want to recover $\bf{\dot{x}}$ from the time series of $\bf{x}$. Any insight on this problem is highly appreciated. Thank you.

$\endgroup$
  • $\begingroup$ The stochastic process that solves (1) is called multivariate Ornstein-Uhlenbeck process. Its sample paths are in general not differentiable. For special choices of $Q$ and the correlation matrix of $\eta$ you can make certain components of the process differentiable a certain number of times. $\endgroup$ – S.Surace Jan 1 '18 at 16:11
1
$\begingroup$

For the general question of estimating a derivative from a noisy time series, there exists a fairly large literature; the best tool is probably determined by which modeling assumptions you want to make about the data.

For example, a Savitzky-Golay filter is a technique for smoothing data that also provides an analytic, closed-form estimate for the derivative.

$\endgroup$
  • $\begingroup$ "the best tool is probably determined by which modeling assumptions you want to make about the data." Right. In particular, for model (1) the derivative is not defined for all components of the process, and applying methods that assume its existence may lead to nonsensical results. $\endgroup$ – S.Surace Jan 1 '18 at 16:22
0
$\begingroup$

to recover $\dot{\mathbf{x}}$ from $\mathbf{x}$, maybe one way is to first get an estimation of $Q$ and $\eta$ (treat $Q$ as a random variable, independent of all $\eta$). Assume an estimation of $Q$ is $\hat{Q}$, and by simple algebra, $<\dot{\mathbf{x}} \dot{\mathbf{x}}^T> = <(\hat{Q}\mathbf{x})(\hat{Q}\mathbf{x})^T> + var[\eta]$.

Probably the simplest method is to apply linear regression -- your output data is $\dot{\mathbf{x}} \approx \frac{\mathbf{x}(t+\Delta t) - \mathbf{x}(t)}{\Delta t}$, input data is $\mathbf{x}$, and your error is $\eta$.

$\endgroup$
  • $\begingroup$ I already tried $\frac{x(t+\delta t)-x(t)}{\delta t}$ and get $<\dot{x}\dot{x}^T>$ but it's not giving me the right answer. $\endgroup$ – Bingkat Nov 2 '17 at 9:50
  • $\begingroup$ I think you cannot directly write $\dot{\mathbf{x}} = \frac{\mathbf{x}(t+\Delta t) - \mathbf{x}(t)}{\Delta t}$. Instead you should do: 1. Applying linear (or nonlinear) regression of $\frac{\mathbf{x}(t+\Delta t) - \mathbf{x}(t)}{\Delta t}$ over $\mathbf{x}$; 2. Get an estimation of $\hat{Q}$ by step 1, and calculate $var [\eta]$ (square sum of residues); 3. Calculate $<\dot{\mathbf{x}} \dot{\mathbf{x}}^T>$ using information in step 2. $\endgroup$ – Xige Yang Nov 3 '17 at 15:30
  • $\begingroup$ Also here, one should first ask whether the difference quotient makes sense for the model in question. $\endgroup$ – S.Surace Jan 1 '18 at 16:30
0
$\begingroup$

Note that $\newcommand\mean[1]{\left\langle #1 \right\rangle} \mean{(-Qx+\eta)(-Qx+\eta)^\top}=\mean{Qxx^\top Q^\top}+\mean{\eta \eta^\top}=\mean{Qxx^\top Q^\top}+B\, \delta(0)$, where $B$ is the covariance matrix of the noise (identity matrix if they are normalized uncorrelated white noise).

The reason that you are having a hard time fitting with the time series is that the second term is proportional to the Dirac $\delta$-function, while the first term is finite. As you take smaller and smaller $\Delta t$s for numerically calculating the derivative the second term grows larger with a factor $(\Delta t)^{-1}$.

As suggested by Xige Yang, you need to apply a linear regression on $\frac{x(t+\Delta t)-x(t)}{\Delta t}$ to find $Q$. Then, $B$ is given by the covariance of the error multiplied by $\Delta t$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.