I am trying to find the stationary distribution of an AR(1) process, where the shock terms themselves are an AR(1) process. That is, the process moves subject to the following 2 equations: \begin{equation} z_{t+1} = \rho z_t + \sigma_t \eta_{z,t+1} \end{equation} \begin{equation} \sigma^2_{t+1} = \max\left\{\nu \sigma_t^2 + \sigma \eta_{\sigma,t+1},\: 0 \right\} \end{equation} where $\eta_{\sigma, t+1}, \eta_{z,t+1} \sim N(0,1)$ are IID.

I believe that the stationary distribution of $\sigma_t \eta_{z,t+1}$ is a linear combination of $\chi$-square distributions. (Based on answers here)

Does this mean that the stationary distribution of $z_{t+1}$ is also the same linear combination of $\chi$-square distributions?

  • $\begingroup$ This looks more like some kind of stochastic volatility shocks rather than AR(1) shocks. $\endgroup$ – Richard Hardy Sep 18 '19 at 6:17
  • $\begingroup$ Yeah I'm trying to prove an existence result for the Bansal and Yaron stochastic volatility model. Does that change the nature of the problem? $\endgroup$ – Flintro Sep 18 '19 at 6:44
  • $\begingroup$ I do not think altering the name changes the nature of the problem, but it changes the relevance of your title (currently a little misleading). $\endgroup$ – Richard Hardy Sep 18 '19 at 7:23

Your Answer

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

Browse other questions tagged or ask your own question.