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.
Bagged decision-tree ensemble for nonlinear macro prediction and feature-importance diagnostics.
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.
The central statistical result behind random forests is that averaging estimators reduces variance by a factor that depends on their pairwise correlation. When estimators are independent, averaging achieves the full variance reduction. When they are positively correlated (as bagged trees are), the reduction is limited by the average correlation. Random forests lower this correlation through feature subsampling.
Breiman (2001) formalized this using the decomposition of ensemble variance into a correlation component and an individual-tree variance component. The forest's prediction variance at a point is , where is the average pairwise correlation between tree predictions and is the variance of a single tree. The first term is irreducible by adding more trees; only reducing through feature subsampling can lower it.
The random forest prediction is the average of individual tree predictions. For classification, this is a majority vote.
Variance of the ensemble prediction expressed as the sum of all pairwise covariances between tree predictions.
Breiman's variance decomposition. The first term persists regardless of . The second term vanishes as . Reducing through feature subsampling directly lowers the irreducible first term.
The asymptotic variance of an infinite forest. This is strictly positive whenever trees are correlated (), which is always the case for bagged trees. Reducing mtry reduces and pushes this floor lower.
Why not set mtry = 1?
At , each split is chosen from a single randomly selected feature. Trees become nearly independent () and ensemble variance approaches , which is excellent. But individual trees are so poor that bias dominates. A tree that can only split on random single features is essentially a random partition of the space. The ensemble of random partitions converges to something useful but requires thousands of trees and large samples. The default provides a practical sweet spot.
Finite-B practice
In practice, the second term becomes negligible by to for most macro applications. The OOB error curve flattens visibly at this point. Growing more trees never hurts prediction accuracy (the forest prediction is a running average), but it does increase computation and memory. The convergence diagnostic is the OOB error curve itself: stop when the curve is flat.
The plotted object solves an equilibrium system at each parameter vector. A path would require an additional law of motion, such as . Without that law, the model identifies endpoints but not the adjustment between them.
Read a parameter move as a new equilibrium or schedule position. The segment between two saved scenarios is not a transition law.
Do not infer speed, overshooting, or smooth adjustment from a static schedule shift.
A point or schedule is solved at the current parameter vector. It is not dated unless the model adds a transition law.
The parameter changes the solved object immediately in the diagram; the diagram does not say how institutions implement that change.
Proof lineage
Typed relationships across modelsParents -- models this proof draws from