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

ARDL / bounds testing Model

Autoregressive distributed lag model for short-run dynamics and long-run relationships among variables that may have mixed integration orders.

概要グラフ数式参考資料

Is there a long-run equilibrium relationship between these variables when I am not sure whether they are I(0), I(1), or a mix?

Background

Cointegration analysis traditionally requires that all variables be integrated of the same order. The Johansen (1991) procedure assumes a pure I(1) system; the Engle-Granger (1987) two-step method requires the same. In applied macro, this precondition is rarely met cleanly. Unit-root tests have limited power, some series appear borderline I(0)/I(1), and pre-testing for integration order introduces its own specification risk. Pesaran, Shin, and Smith (2001) - hereafter PSS - proposed the bounds-testing approach: estimate an ARDL model in error-correction form and test for the significance of the lagged-level terms. The test provides two sets of critical values - one assuming all regressors are I(0), one assuming all are I(1) - and the sample F-statistic is compared to both. If it exceeds the upper bound, cointegration is confirmed regardless of integration order. If it falls below the lower bound, no cointegration. If it falls between the bounds, the test is inconclusive.

The ARDL model itself is old - Hendry and von Ungern-Sternberg (1981) used distributed-lag specifications throughout the 1970s and 1980s. What PSS contributed was an inferential framework that works in the I(0)/I(1)I(0)/I(1)I(0)/I(1) gray zone. The model reparameterizes the standard ARDL(p,q1,…,qk)\text{ARDL}(p, q_1, \ldots, q_k)ARDL(p,q1​,…,qk​) into a conditional error-correction form where the long-run coefficients appear explicitly as ratios of lagged-level coefficients and the short-run dynamics appear as differenced terms.

Estimation is straightforward OLS after selecting the lag orders (p,q1,…,qk)(p, q_1, \ldots, q_k)(p,q1​,…,qk​) by information criteria. The bounds test is an F-test on the joint significance of the lagged levels of yty_tyt​ and xtx_txt​ in the error-correction equation. Pesaran and Shin (1999) showed that the ARDL approach produces consistent estimates of the long-run parameters even when the regressors are endogenous, provided the lag orders are correctly specified.

The ARDL bounds test is the workhorse cointegration test in applied development economics, energy economics, and agricultural economics. Google Scholar returns tens of thousands of citations to PSS (2001), making it one of the most cited econometrics papers ever. It is the default approach in EViews, Stata's ardl command, and the R package dynamac.

Connections to explore

Math ↗The structure and its conditions, step by step.References ↗Academic sources and guidance for using the model.

How the Parts Fit Together

πy\pi_yπy​

Error-correction speed

Coefficient on the lagged level of yty_tyt​. Must be negative for stability. Its absolute value measures how fast disequilibrium is corrected per period.

The ARDL(p,q1,…,qk)\text{ARDL}(p, q_1, \ldots, q_k)ARDL(p,q1​,…,qk​) model takes a single dependent variable yty_tyt​ and kkk regressors x1t,…,xktx_{1t}, \ldots, x_{kt}x1t​,…,xkt​. Each regressor enters with its own lag order qjq_jqj​, selected by AIC, BIC, or similar criteria. The model includes ppp lags of yty_tyt​ and up to qjq_jqj​ lags of each xjtx_{jt}xjt​. Deterministic components (intercept, restricted trend, unrestricted trend) are included depending on the case (PSS tabulated five cases).

The error-correction reparameterization rewrites the ARDL as Δyt=α0+πyyt−1+∑jπjxj,t−1+(short-run differenced terms)+ut\Delta y_t = \alpha_0 + \pi_y y_{t-1} + \sum_j \pi_j x_{j,t-1} + (\text{short-run differenced terms}) + u_tΔyt​=α0​+πy​yt−1​+∑j​πj​xj,t−1​+(short-run differenced terms)+ut​. The coefficient πy\pi_yπy​ is the speed of adjustment: it must be negative for the ECM to be stable. The long-run coefficients are recovered as βj=−πj/πy\beta_j = -\pi_j / \pi_yβj​=−πj​/πy​. The bounds test is an F-test for H0:πy=π1=⋯=πk=0H_0: \pi_y = \pi_1 = \cdots = \pi_k = 0H0​:πy​=π1​=⋯=πk​=0 (no levels relationship).

Output includes: (1) the estimated long-run coefficients βj\beta_jβj​ with standard errors (the delta method or Bewley transform), (2) the short-run coefficients on the differenced terms, (3) the error-correction speed πy\pi_yπy​, (4) the bounds F-statistic and its position relative to the I(0)I(0)I(0) and I(1)I(1)I(1) critical value bounds, and (5) diagnostic tests (serial correlation, heteroskedasticity, functional form, normality).

Assumptions

Testable

No I(2) or higher variables

All variables are I(0) or I(1). The bounds test is invalid for I(2) processes.

If violated:I(2) variables make the asymptotic theory invalid. The bounds critical values no longer bracket the true distribution.

Testable

Correct lag specification

The ARDL lag orders (p,q1,…,qk)(p, q_1, \ldots, q_k)(p,q1​,…,qk​) are chosen to eliminate serial correlation in the residuals.

If violated:Under-specified lags produce serially correlated residuals, biasing the bounds F-statistic upward and inflating the cointegration rejection rate.

Maintained

Single cointegrating relationship

There is at most one long-run relationship with yty_tyt​ as the dependent variable.

If violated:If multiple cointegrating vectors exist, the ARDL identifies only the one normalized on y. The others are ignored, potentially biasing the estimated long-run coefficients.

Testable

Weak exogeneity of regressors

The regressors xjtx_{jt}xjt​ do not respond to the error-correction term in the long run (or the lag orders are sufficient to proxy any endogeneity).

If violated:If xjtx_{jt}xjt​ is strongly endogenous and lag orders are insufficient, the OLS estimates are inconsistent. Pesaran and Shin (1999) showed that sufficient lags restore consistency.

Testable

No structural breaks

The long-run and short-run parameters are constant over the sample.

If violated:Structural breaks can create spurious cointegration results. CUSUM and CUSUMSQ tests are standard post-estimation diagnostics.

Testable

Innovation regularity

utu_tut​ is i.i.d. with zero mean and finite variance.

If violated:Heteroskedasticity does not bias the point estimates but invalidates standard inference; use heteroskedasticity-robust standard errors. (The Kripfganz-Schneider 2020 critical values are finite-sample and deterministic-case adjustments, not a heteroskedasticity correction.)

Applications

The World Bank and IMF use ARDL bounds testing to study long-run determinants of economic growth, financial development, and energy consumption in developing countries where time series are short (30 to 60 observations) and integration orders are uncertain. The single-equation framework is attractive because it avoids the Johansen procedure's requirement for a well-specified VAR with sufficient degrees of freedom.

Energy economists use the ARDL to test the Environmental Kuznets Curve (EKC) hypothesis: whether pollution first rises and then falls with per-capita income. The long-run coefficient on income-squared must be negative and significant. The bounds test confirms whether a genuine long-run relationship exists or whether the quadratic pattern is spurious.

The ARDL cannot handle more than one cointegrating vector in a single equation. When the system has multiple long-run relationships (e.g., a money-demand equation and a Phillips curve that share variables), the Johansen VECM is needed. The ARDL also requires that no variable is I(2)I(2)I(2), which rules out nominal price series in hyperinflationary episodes. With very short samples (T<30)(T < 30)(T<30), the bounds F-statistic has low power and the critical values from PSS (2001) may not apply - Narayan (2005) tabulated small-sample critical values.

Extensions include the nonlinear ARDL (NARDL) of Shin, Yu, and Greenwood-Nimmo (2014), which decomposes regressors into positive and negative partial sums to test for asymmetric long-run and short-run effects. The quantile ARDL (QARDL) extends the framework to quantile regression. Panel ARDL estimators (Pesaran, Shin, and Smith 1999) handle cross-country heterogeneity through mean-group and pooled mean-group estimators.

Extensions and open questions

Variants

  • NARDL - nonlinear extension with asymmetric positive/negative decomposition
  • QARDL - quantile autoregressive distributed lag
  • Panel ARDL - cross-country extension (MG, PMG, DFE estimators)
  • Bootstrap ARDL - McNown et al. augmented bounds test with bootstrap p-values
  • Structural break ARDL - with regime-shift dummies and break-date estimation

Open questions

  • The bounds test is inconclusive when the F-statistic falls between the I(0) and I(1) bounds, which happens frequently in practice. Practitioners often report the result as 'cointegration found' if above the upper bound but have no clear recommendation in the inconclusive zone.
  • The PSS (2001) critical values assume strict exogeneity of the forcing variables in the long run. When regressors are endogenous, the bootstrapped bounds test of McNown et al. (2018) is more appropriate but less widely used.
  • Sensitivity of long-run estimates to lag-order selection remains a practical concern. Different information criteria (AIC vs BIC vs HQ) can select different lag structures that produce materially different long-run coefficients.

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