メインコンテンツへ移動

Macro by Mark の Cookie.ログイン、言語、安全性のために必須 Cookie を使用します。許可いただいた場合は、設定の保存、分析、Meta(Facebook)ピクセルによる広告計測のために任意の Cookie も使用します。詳しくはCookie ポリシーをご覧ください。

Macro by MarkMacro by Mark
  • ホーム
読み込み中…
読み込み中…
← モデル
Macro概念政策モデル

Bayesian VAR (BVAR) Model

Bayesian shrinkage on a vector autoregression - Minnesota-style priors stabilize forecasts for medium-sized macro systems.

概要グラフ数式参考資料

Graph workspace

Shrinkage vs. forecast variance tradeoff with IRF overlay

Graph workspace

The defining feature of BVAR is the bias-variance tradeoff controlled by the shrinkage hyperparameter λ\lambdaλ. Showing how IRFs and forecast intervals change as λ\lambdaλ moves from 0 (pure prior) to ∞\infty∞ (pure OLS) is the most direct way to build intuition for what shrinkage actually does.

Chart axis metadata
xHorizonquarters · 0-20yResponse / Forecastpercentage points · -2 to 2

Shaded ranges come from assumed parameters in a simulated example. They are not estimated from observed data.

Own response under the posterior static chart preview showing Own response under the posterior
Cross response under the posterior static chart preview showing Cross response under the posterior
Own response under the OLS estimate static chart preview showing Own response under the OLS estimate
xHorizonquarters · 0-20yResponse / Forecastpercentage points · -2 to 2

Current results

Posterior own-lag

Explain: Posterior own-lag
a^11Bayes=V0−1⋅1+T⋅a11OLSV0−1+T\hat{a}_{11}^{\text{Bayes}} = \frac{V_0^{-1} \cdot 1 + T \cdot a_{11}^{\text{OLS}}}{V_0^{-1} + T}a^11Bayes​=V0−1​+TV0−1​⋅1+T⋅a11OLS​​

The shrinkage-weighted compromise between the prior belief (a random walk, coefficient 1) and the OLS point estimate. Closer to 1 under tight shrinkage, closer to OLS under loose.

0.9000

coefficient

Posterior cross-lag

Explain: Posterior cross-lag
a^12Bayes=V0,cross−1⋅0+T⋅a12OLSV0,cross−1+T\hat{a}_{12}^{\text{Bayes}} = \frac{V_{0,\text{cross}}^{-1} \cdot 0 + T \cdot a_{12}^{\text{OLS}}}{V_{0,\text{cross}}^{-1} + T}a^12Bayes​=V0,cross−1​+TV0,cross−1​⋅0+T⋅a12OLS​​

The shrinkage-weighted cross-variable coefficient. Prior mean is 0, so this is pulled toward zero more aggressively than the own lag.

0.0375

coefficient

IRF half-life

Explain: IRF half-life
h∗=min⁡{h:∣IRF(h)∣<0.5}h^* = \min\{h : |\text{IRF}(h)| < 0.5\}h∗=min{h:∣IRF(h)∣<0.5}

Number of quarters until the impulse response to a unit shock in variable 1 decays below 50% of its impact value, counted on the posterior system even when that lies beyond the 24 plotted quarters; blank when the response never halves (an explosive or unit-root posterior). Longer under tight shrinkage: the Minnesota prior centers the own lag on a random walk (coefficient 1), so tightening it moves the posterior toward a unit root and slows the decay.

7.00

quarters

4Q forecast interval width

Explain: 4Q forecast interval width
CI4=2×1.0×σ1∑j=03(Φj)112\text{CI}_{4} = 2 \times 1.0 \times \sigma_1 \sqrt{\sum_{j=0}^{3} (\boldsymbol{\Phi}_j)_{11}^{2}}CI4​=2×1.0×σ1​∑j=03​(Φj​)112​​

Width of the 68% forecast band (plus or minus one standard deviation) for variable 1 at the one-year horizon, from variable 1's own shocks carried through the posterior-mean coefficients; parameter uncertainty is not included. Because the Minnesota prior pulls the own lag toward a random walk, tighter shrinkage makes shocks persist and WIDENS this band. In a full BVAR the predictive band also carries parameter uncertainty, which a tighter prior reduces, so there the net effect can go either way.

3.47 pp

percentage points

Controls

Balanced blend
Explain: Looseness of the starting beliefs

Overall shrinkage: 0.2

Controls how strongly the estimates are pulled toward the model's starting beliefs.

Low λ\lambdaλ: the IRF flattens toward the random-walk prior, its half-life lengthens and the one-year forecast band widens, because near-random-walk dynamics let shocks accumulate. High λ\lambdaλ: the IRF approaches the OLS estimate, decays faster and the band narrows.

Weak cross-links
Explain: How much other variables count

Cross-variable shrinkage: 0.5

How much weight other variables' past values get compared with a variable's own past.

Low values flatten cross-variable IRF responses toward zero. High values (near 1) treat own and cross lags symmetrically.

Long memory
Explain: Data-only estimate for variable 1

OLS own-lag (var 1): 0.8

What the raw data says about how much variable 1 repeats itself, before shrinkage.

Determines the OLS IRF decay rate (dashed gray). The posterior IRF is pulled between this and the prior mean of 1.

Pushes the other up
Explain: Variable 2's effect on 1

OLS cross-lag (2 → 1): 0.15

What the raw data says about how much variable 2 affects variable 1.

Affects the shape of the IRF at intermediate horizons. Shrinkage pulls this toward the prior mean of 0.

1
Explain: Shock size for variable 1

How large the random shocks to variable 1 are, which scales forecast uncertainty.

Wider σ1\sigma_1σ1​ widens the one-year forecast band in proportion. The impulse responses are to a unit shock, so their shape does not change.

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
ヘルプ
プライバシーポリシー利用規約倫理とコンプライアンス

© 2026 Mark Jayson Nation LLC