Skip to content
Get started

Steady-state gain of a fair-value filter

A fair value follows a random walk, xt=xt−1+wtx_t=x_{t-1}+w_t with Var⁡(wt)=1\operatorname{Var}(w_t)=1. Each step you observe a noisy quote yt=xt+vty_t=x_t+v_t with Var⁡(vt)=1\operatorname{Var}(v_t)=1, all noises independent. You track xtx_t with a Kalman filter, x^t=x^t−1+K(yt−x^t−1)\hat x_t=\hat x_{t-1}+K(y_t-\hat x_{t-1}).

What is the steady-state gain KK? Give a decimal to 3 places.

Show hints (2)+
  1. Write the predicted variance at the next step in terms of this step's prior PP.
  2. Solve the fixed point P=P/(P+1)+1P=P/(P+1)+1, then K=P/(P+1)K=P/(P+1).

Answer

Reveal answer →

0.618 (± 0.002)

Create a free account to see the full worked solution and track your progress.

Interview format: Hedge Fund Quant, Systematic & Data-Driven Trading

Related questions