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

Satellite models

Smaller models that hang off the core macroeconometric system to handle housing, energy, financial conditions, or other special topics.

OverviewGraphMathReferences

Graph workspace

Satellite response fan conditioned on core macro scenario

Graph workspace

The defining visualization for satellite models is how the satellite target variable responds to a macro shock propagated through the core model. Showing the core's GDP path alongside the satellite's response (e.g., housing starts, credit losses) reveals the transmission mechanism and amplification/dampening across the bridge.

Chart axis metadata
xQuarters after macro shockquarters · 0 to 16yLevel or deviation from baselineindex (core GDP) or level (satellite variable) · varies by satellite domain
Core shock and the satellite response
it drives static chart preview showing Core macro variable, Satellite variable, Baseline
xQuarters after macro shockquarters · 0 to 16yLevel or deviation from baselineindex (core GDP) or level (satellite variable) · varies by satellite domain

Current results

Satellite peak deviation

Explain: Satellite peak deviation
y^h∗(s),y^t(s)=100 yt(s)−ybase(s)ybase(s),h∗=arg⁡max⁡t∣y^t(s)∣\hat{y}^{(s)}_{h^*}, \quad \hat{y}^{(s)}_t = 100\,\frac{y_t^{(s)} - y_{\text{base}}^{(s)}}{y_{\text{base}}^{(s)}}, \quad h^* = \arg\max_t |\hat{y}^{(s)}_t|y^​h∗(s)​,y^​t(s)​=100ybase(s)​yt(s)​−ybase(s)​​,h∗=argmaxt​∣y^​t(s)​∣

Largest deviation of the satellite variable from its baseline, in percent of that baseline, with its sign: a decline in every preset. For a housing satellite, this is the peak decline in housing starts. For a credit-loss satellite, read it as the peak increase in losses.

-2.31%

percent deviation from baseline

Transmission lag

Explain: Transmission lag
arg⁡min⁡t(yt(s))−arg⁡min⁡t(gt)\arg\min_t (y_t^{(s)}) - \arg\min_t (g_t)argmint​(yt(s)​)−argmint​(gt​)

How many quarters the satellite trough lags the core GDP trough, followed past the 32 plotted quarters if the trough comes later. Reveals the delay in macro-to-sector transmission. Blank when the satellite never moves (no shock, or β1+δ2/4=0\beta_1 + \delta_2/4 = 0β1​+δ2​/4=0).

3.00

quarters

Amplification ratio

Explain: Amplification ratio
y^h∗(s)  /  ε0g\hat{y}^{(s)}_{h^*} \;/\; \varepsilon_0^gy^​h∗(s)​/ε0g​

Satellite's peak percentage deviation from its baseline divided by the GDP shock in percentage points, which is the core's deepest point. The toy draws the satellite in percent deviation from its baseline, so this is a ratio of percent moves and needs no baseline level. Ratios above 1 mean the satellite amplifies the macro shock; below 1 means it dampens; a negative ratio means it moves against GDP. Blank when there is no shock.

1.15

ratio (dimensionless)

Satellite half-life

Explain: Satellite half-life
min⁡{t>h∗:∣yt(s)−ybase(s)∣<0.5∣yh∗(s)−ybase(s)∣}−h∗,h∗=arg⁡max⁡t∣yt(s)−ybase(s)∣\min\{t > h^* : |y_t^{(s)} - y_{\text{base}}^{(s)}| < 0.5 |y_{h^*}^{(s)} - y_{\text{base}}^{(s)}|\} - h^*, \quad h^* = \arg\max_t |y_t^{(s)} - y_{\text{base}}^{(s)}|min{t>h∗:∣yt(s)​−ybase(s)​∣<0.5∣yh∗(s)​−ybase(s)​∣}−h∗,h∗=argmaxt​∣yt(s)​−ybase(s)​∣

Quarters from the satellite variable's trough until it has recovered halfway, measured from the trough because the satellite starts at its baseline and only reaches its trough later, and followed past the 32 plotted quarters if needed. Longer half-lives indicate sectors that are slow to recover from downturns. Blank when the satellite never moves.

7.00

quarters

Controls

reacts moderately
Explain: How much the sector reacts

Satellite: core macro sensitivity: 0.8

How strongly the sector responds to wider economic conditions; higher values amplify economic shocks.

Higher β1\beta_1β1​ produces a deeper trough in the satellite variable for a given GDP decline. The satellite panel shows larger deviations from baseline.

recovers gradually
Explain: How slowly the sector recovers

Satellite: own persistence: 0.6

How much of last period's sector level carries forward; higher means slower recovery.

Higher β2\beta_2β2​ elongates the satellite's response. The trough is similar but the recovery takes many more quarters.

drawn-out downturn
Explain: How long the downturn lasts

Core: GDP persistence: 0.85

How much GDP growth carries over each quarter; higher values make recessions more persistent.

Higher ρg\rho_gρg​ produces a longer GDP trough in the top panel, which in turn prolongs the satellite's exposure to negative macro conditions.

rate cuts cushion mildly
Explain: How much rate cuts cushion

Bridge: interest rate channel: -0.3

How the policy rate feeds into the sector; more negative means rate cuts cushion the sector.

More negative δ2\delta_2δ2​ partially offsets the GDP hit on the satellite. The satellite trough is shallower because rate cuts dampen the macro impulse.

moderate downturn
Explain: Size of the GDP hit

Core GDP shock size: -2

How big the initial hit to GDP growth is, in percentage points.

Larger shocks produce proportionally deeper responses in both the core and satellite panels (linear system). Very large shocks may push the satellite to implausible values.

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