Skip to main content

Cookies on Macro by Mark.We use essential cookies for sign-in, locale, and security. With your permission, we also use optional cookies for preferences, for analytics, and for advertising measurement with the Meta (Facebook) Pixel. Read our cookie policy.

Macro by MarkMacro by Mark
  • Home
Loading…
Loading…
← Models
MacroConceptsPolicyModels

Kalman filter Model

Recursive state estimator that updates latent macro states as each noisy observation arrives.

OverviewGraphMathReferences

Understand the mathematics

Follow the same equations and assumptions one section at a time or in a continuous reading.

Hover or focus a symbol defined in this model to find its exact matches. Select it to keep the highlight; Escape clears the selection.

State-Space Model Representation

The Kalman filter operates on a linear Gaussian state-space model. We establish the notation and the two equations that define the model. All subsequent derivations follow from these primitives.

αt=Ftαt−1+ct+Gtηt,ηt∼N(0,Qt)\alpha_t = F_t \alpha_{t-1} + c_t + G_t \eta_t, \quad \eta_t \sim N(0, Q_t)αt​=Ft​αt−1​+ct​+Gt​ηt​,ηt​∼N(0,Qt​)
αt\alpha_tαt​FtF_tFt​QtQ_tQt​

The transition (state) equation. The state evolves linearly with additive Gaussian noise. GtG_tGt​ is the state-disturbance loading (selection) matrix - deliberately not written RtR_tRt​, which is reserved for the measurement-noise covariance.

yt=Htαt+dt+εt,εt∼N(0,Rt)y_t = H_t \alpha_t + d_t + \varepsilon_t, \quad \varepsilon_t \sim N(0, R_t)yt​=Ht​αt​+dt​+εt​,εt​∼N(0,Rt​)
HtH_tHt​αt\alpha_tαt​RtR_tRt​

The measurement (observation) equation. Observations are linear functions of the state plus noise with covariance RtR_tRt​, matching the notation used in the graph tab.

E[ηtεs′]=0∀t,s;E[ηtηs′]=Qtδts;E[εtεs′]=RtδtsE[\eta_t \varepsilon_s^\prime] = 0 \quad \forall t, s; \quad E[\eta_t \eta_s^\prime] = Q_t \delta_{ts}; \quad E[\varepsilon_t \varepsilon_s^\prime] = R_t \delta_{ts}E[ηt​εs′​]=0∀t,s;E[ηt​ηs′​]=Qt​δts​;E[εt​εs′​]=Rt​δts​
QtQ_tQt​RtR_tRt​

State and measurement disturbances are mutually independent and serially uncorrelated.

α0∼N(a0,P0),independent of {ηt},{εt}\alpha_0 \sim N(a_0, P_0), \quad \text{independent of } \{\eta_t\}, \{\varepsilon_t\}α0​∼N(a0​,P0​),independent of {ηt​},{εt​}
α0\alpha_0α0​

The initial state has a known (or diffuse) Gaussian distribution.

Transition checks

Sampled dynamics
zt=G(zt−1;θ)+ηt,yt=Hzt+εtz_t=G(z_{t-1};\theta)+\eta_t,\qquad y_t=H z_t+\varepsilon_tzt​=G(zt−1​;θ)+ηt​,yt​=Hzt​+εt​

The state is updated at observed dates. A line between dates is a display convention unless the model specifies behavior inside the interval.

Read adjacent points as adjacent sample calculations. Smoothness is a property of the estimator, not of the chart line.

Do not infer within-period behavior from a sampled estimator.

Variable timing

Latent state

zt=G(zt−1;θ)+ηtz_t=G(z_{t-1};\theta)+\eta_tzt​=G(zt−1​;θ)+ηt​

The state updates at sample dates. The visible line is only as smooth as the estimated state sequence.

Observation equation

yt=Hzt+εty_t=Hz_t+\varepsilon_tyt​=Hzt​+εt​

Observed series can move differently from latent states because measurement noise and loadings matter.

Keep going

Read next

Continue with a related concept, model, or example.

Theory model

IS-LM Model

a well-connected theoretical model two steps from here

New angle: theoretical model →
Macro by Mark
Help
Privacy PolicyTerms of UseEthics and Compliance

© 2026 Mark Jayson Nation LLC