Abstract
This paper proposes residual-based tests for cointegration with three-regime threshold autoregressive (TAR) adjustment. We propose Wald-type and \(t\)-type tests that have the null hypothesis of linear no cointegration and the alternative of cointegration with three-regime TAR adjustment and also derive the asymptotic distributions. Monte Carlo simulations show that the proposed tests perform better than the Engle–Granger cointegration test and the cointegration test in a two-regime TAR model introduced by Enders and Siklos (J Bus Econ Stat 19:166–176, 2001), under cointegration with three-regime TAR adjustment, particularly when the threshold and sample size increase. When we apply these tests to the money demand of the U.S., the proposed tests reject the null of no cointegration whereas other tests do not.
Similar content being viewed by others
Notes
Park and Shintani (2010) developed unit root tests in various transitional AR models, including the three-regime TAR model. Maki (2009) reviewed unit root tests in three-regime TAR models and investigated the power of these models. Other unit root tests in nonlinear frameworks have also been proposed by Enders and Granger (1998), Caner and Hansen (2001), Kapetanios et al. (2003), and Bec et al. (2010). Choi and Moh (2007) investigated the power of some unit root tests in nonlinear frameworks.
If (2) is a three-regime TAR model with a symmetric adjustment, the model under the alternative is \(-1<\phi _1=\phi _2<1\). In addition, a special case of a symmetric three-regime TAR model is given by \(u_t=\phi u_{t-1}\mathbf {1}_{\{|u_{t-1}| > \lambda \}}+e_t\). Since (2) also includes these restricted models, we only consider general model (2).
Although it is possible that \(\Delta \hat{u}_{t-j}\) also follows a TAR process, for the sake of simplicity, we do not consider this case as in Enders and Siklos (2001). Even if \(\Delta \hat{u}_{t-j}\) is a TAR process, the asymptotic distribution of the test statistic does not change.
If \(\tilde{\lambda }_1=\tilde{\lambda }_2=0\) or \(\tilde{\lambda }_1=\tilde{\lambda }_2\), the test statistic is found to be similar to the \(\varPhi \) statistic by using an \(F\) statistic of Enders and Siklos (2001), although they did not show the asymptotic distribution.
Asymptotic distributions of the tests under the assumption that the threshold parameter degenerates are obtained by the replacing the indicator functions from \((1a)\) to \((1d)\) in Theorem 1 with \(\mathbf {1}\{ W^{*} \le 0 \}\) or \(\mathbf {1}\{ W^{*} > 0 \}\). Details of the asymptotic distributions and critical values are available with the author and will be provided on request.
Owing to space constraints, we have not tabulated finite size critical values. These are available with the author and will be provided on request.
Critical values of the ES-\(\varPhi \) test are 6.208, 7.272, and 9.640 at 90, 95, and 99 %, respectively.
In fact, homoskedasticity tests reject the null hypothesis of homoskedasticity of the residual variance. The results are available with the author and will be provided on request.
References
Andrews DWK, Ploberger H (1994) Optimal tests when a nuisance parameter is present only under the alternative. Econometrica 62:1383–1414
Balke NS, Fomby T (1997) Threshold cointegration. Int Econ Rev 38:627–643
Bec F, Ben Salem M (2004) Tests for unit-root versus threshold specification with an application to the purchasing power parity relationship. J Bus Econ Stat 22:382–395
Bec F, Ben Salem M (2010) Detecting mean reversion in real exchange rates from a multiple regime STAR model. Annal Eco Stat 99(100):395–428
Bec F, Guay A, Guerre E (2008) Adaptive consistent unit-root tests based on autoregressive threshold model. J Econom 142:94–133
Brockwell PJ, Liu J, Tweedie R (1992) On the existence of stationary threshold autoregressive moving average processes. J Time Ser Anal 13:95–107
Caner M, Hansen BE (2001) Threshold autoregression with a unit root. Econometrica 69:1555–1596
Chan KS, Petrucelli J, Tong H, Woolford S (1985) A multiple threshold AR(1) model. J Appl Prob 22:267–279
Chan KS, Tong H (1985) On the use of the deterministic Lyapunov functions for the egodicity of stochastic difference equations. Adv Appl Prob 17:666–678
Choi CY, Moh YK (2007) How useful are tests for unit-root in distinguishing unit-root processes from stationary but nonlinear processes? Econom J 10:82–112
Clements MP, Galvãro AB (2003) Testing the expectations theory of the term structure of interest rates in threshold models. Macroecon Dyn 7:567–585
Cuthbertson K, Taylor MP (1987) The demand for money: a dynamic rational expectation model. Econ J 97:65–76
Davies RB (1987) Hypothesis testing when a nuisance parameter is present under the alternative. Biometrika 74:33–43
Enders W, Granger CWJ (1998) Unit root tests and asymmetric adjustment with an example using the term structure of interest rates. J Bus Econ Stat 16:304–311
Enders W, Siklos PL (2001) Cointegration and threshold adjustment. J Bus Econ Stat 19:166–176
Engle RF, Granger CWJ (1987) Cointegration and error correction: representation, estimation, and testing. Econometrica 55:251–276
Hansen BE (1996) Inference when a nuisance parameter is not identified under the null hypothesis. Econometrica 64:414–430
Hansen BE, Seo B (2002) Testing for two-regime threshold cointegration in vector error correction models. J Econom 110:293–318
Johansen S (1991) Estimation and hypothesis testing of cointegration vectors in Gaussian vector autoregressive models. Econometrica 59:1551–1580
Kapetanios G, Shin Y (2006) Unit root tests in three-regime SETAR models. Econom J 9:252–278
Kapetanios G, Shin Y, Snell A (2003) Testing for a unit root in the nonlinear STAR framework. J Econom 112:359–379
Kapetanios G, Shin Y, Snell A (2006) Testing for cointegration in nonlinear smooth transition error correction models. Econom Theory 22:279–303
Kiliç R (2011) Testing for co-integration and nonlinear adjustment in a smooth transition error correction model. J Time Ser Anal 32:647–660
Kurtz TG, Protter P (1991) Weak limit theorems for stochastic integrals and stochastic difference equations. Annal Prob 19:1035–70
Lo MC, Zivot E (2001) Threshold cointegration and nonlinear adjustment to the low of one price. Macroecon Dyn 5:533–576
Maki D (2009) Tests for a unit root using three-regime TAR models: power comparison and some applications. Econom Rev 28:335–363
Maki D, Kitasaka S (2006) The equilibrium relationship among money, income, prices, and interest rates: evidence from a threshold cointegration test. Appl Econ 38:1585–1592
Milbourne R (1987) Re-examining the Buffer-Stock model of money. Econ J 97:130–142
Mizen P (1997) Microfoundations for a stable demand for money function. Econ J 107:1202–1212
Park JY, Shintani M (2010) Testing for a unit root test against transitional autoregressive models. Working paper No. 05–W10, Vanderbilt University
Pippenger MK, Goering GE (1993) A note on the empirical power of unit root tests under threshold processes. Oxf Bull Econ Stat 55:473–481
Pippenger MK, Goering GE (2001) Additional results on the power of unit root and cointegration tests under threshold processes. Appl Econ Lett 7:641–644
Phillips PCB, Ouliaris S (1990) Asymptotic properties of residual based tests for cointegration. Econometrica 58:165–193
Sollis R, Leybourne S, Newbold P (2002) Tests for symmetric and asymmetric nonlinear mean reversion in real exchange rates. J Money Credit Bank 34:686–700
Sarno L (1999) Adjustment costs and nonlinear dynamics in the demand for money: Italy, 1861–1991. Int J Econ Financ 4:155–177
Sarno L, Taylor MP, Peel DA (2003) Nonlinear equilibrium correction in U.S. real money balances, 1869–1997. J Money Credit Bank 35:787–797
Seo M (2006) Bootstrap testing for the null of no cointegration in a threshold vector error correction model. J Econom 134:129–150
Seo M (2008) Unit root test in a threshold autoregression: asymptotic theory and residual-based bootstrap. J Money Credit Bank 24:1699–1716
Taylor AM (2001) Potential pitfalls for the purchasing-power-parity puzzle? Sampling and specification biases in mean-reversion tests of the law of one price. Econometrica 69:473–498
Acknowledgments
We would like to thank Tomoyoshi Yabu and the seminar participants at University of the Ryukyus for helpful suggestions and comments. This research was supported by KAKENHI (Grant No: 23730220, 25380272, 24530375).
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1
Proof of Theorem 1
From (21), we obtain that \(T^{-1/2}\ell _{11}^{-1}\hat{u}_t \Rightarrow W^{*}(r)\). We have
where \(\bar{\lambda }_1=\lim _{T \rightarrow \infty } T^{-1/2}\lambda _1\) and \(\tilde{\lambda }_1=\ell _{11}^{-1} \bar{\lambda }_{1}\). From \(\epsilon _t=\sum _{j=0}^{\infty }D_j\xi _{t-j}^{\prime }\eta =D(L)\xi _t^{\prime }\eta \) and Lemma 2.1 of Phillips and Ouliaris (1990), we have
where \(D(1)=\sum _{j=0}^{\infty }D_j\). (38), (39), and the continuous mapping theorem (CMT) and Theorem 2.2 of Kurtz and Protter (1991) yield
and
A similar type of analysis can be applied to the term for \(\mathbf {1} \{ \hat{u}_{t-1} > \lambda _2 \}\). Therefore, it can be also shown that
and
Proof of Theorem 2
Part of (2a) The statistic \(W_T(\lambda )\) can be written as
where
and \(\mathbf {Q}_p=\mathbf {I}-\mathbf {M}_p(\mathbf {M}_p^{\prime }\mathbf {M}_p)^{-1} \mathbf {M}_p^{\prime }\) is a \(T \times T\) idempotent matrix. \(\mathbf {M}_p\) is the matrix of observations on \(\Delta \hat{u}_{t-p}^p\). Since under \(H_0\) with \(\rho _1=\rho _2=0\), \(\hat{\rho }\) is given by \(\hat{\rho }=(\mathbf {U}^{\prime } \mathbf {Q}_p \mathbf {U})^{-1} \mathbf {U}^{\prime } \mathbf {Q}_p \varvec{\epsilon }\), where \(\varvec{\epsilon }=(\epsilon _1,\ldots , \epsilon _T)^{\prime }\), (44) is expressed by
Let \(\mathbf {U}_1=(\hat{u}_0 \mathbf {1}\{\hat{u}_0 \le \lambda _1 \}, \ldots , \hat{u}_{T-1} \mathbf {1}\{\hat{u}_{T-1} \le \lambda _1 \} )^{\prime }\) and \(\mathbf {U}_2=(\hat{u}_0 \mathbf {1}\{\hat{u}_0 > \lambda _2 \}, \ldots , \hat{u}_{T-1} \mathbf {1}\{\hat{u}_{T-1} > \lambda _2 \} )^{\prime }\). From the orthogonality between \(\mathbf {1} \{ \hat{u}_{t-1} \le \lambda _1 \}\) and \(\mathbf {1} \{ \hat{u}_{t-1}>\lambda _2 \}\), we can decompose (45) as
Consider \(\varvec{\epsilon }^{\prime } \mathbf {Q}_p \mathbf {U}_1(\mathbf {U}_1^{\prime } \mathbf {Q}_p \mathbf {U}_1)^{-1} \mathbf {U}_1^{\prime } \mathbf {Q}_p \varvec{\epsilon }\). It follows from CMT that \(T^{-1}\mathbf {U}_1^{\prime }\mathbf {M}_p=O_p(1)\), \(T^{-1}\mathbf {M}_p^{\prime }\mathbf {M}_p=O_p(1)\), and \(T^{-1/2}\mathbf {M}_p^{\prime }\varvec{\epsilon }=O_p(1)\). Combining these results and Theorem 1, we have
and
Similarly, it can be shown that
and
Next, we consider the variance \(\hat{\sigma }^2\). Note that \(\hat{\rho }_1=O_p(T^{-1}),\, \hat{\rho }_2=O_p(T^{-1})\), and \((\hat{\alpha }_j-\alpha _j)=O_p(T^{-1/2})\). Then, using Lemma 2.2 of Phillips and Ouliaris (1990), we obtain
We can deduce the required results from the asymptotic distribution of \(W_T(\lambda )\) and Assumption 2.
Under the alternative hypothesis, \(W_T(\lambda )\) can be written as
Note that under the alternative hypothesis, \(T^{-1}\mathbf {U}^{\prime }\mathbf {Q}_p \mathbf {U}=O_p(1)\), \(T^{-1/2}\mathbf {U}^{\prime } \mathbf {Q}_p \varvec{\epsilon }=O_p(1)\), and \(\hat{\sigma }^2=O_p(1)\). We can see that
Therefore, \(W_T(\lambda )\) diverges to infinity as \(T \rightarrow \infty \). This also implies that the test statistic diverges to infinity as \(T \rightarrow \infty \).
Part of (2b) From the proof of part (2a), the statistic \(t_T(\lambda )_{\max }\) can be written as
Since \(\hat{\rho }_i\,(i=1,2)\) is given by \(\hat{\rho }_i=(\mathbf {U}_i^{\prime } \mathbf {Q}_p \mathbf {U}_i)^{-1} \mathbf {U}_i^{\prime } \mathbf {Q}_p \varvec{\epsilon }\) under \(H_0\) with \(\rho _1=\rho _2=0,\, t_T(\lambda )_{\max }\) is expressed as
Using from (46) to (50), we have
We can deduce the required results from the asymptotic distribution of \(t_T(\lambda )_{\max }\) and Assumption 2. Next, we consider the properties of the test statistic under the alternative hypothesis. Under the alternative hypothesis, \(t_T(\lambda )_{\max }\) can be written as
Then, we have
from \(T^{-1} \mathbf {U}_1^{\prime } \mathbf {Q}_p \mathbf {U}_1=O_p(1),\, T^{-1/2} \mathbf {U}_1^{\prime }\mathbf {Q}_p \varvec{\epsilon }=O_p(1),\, T^{-1} \mathbf {U}_2^{\prime } \mathbf {Q}_p \mathbf {U}_2=O_p(1), T^{-1/2} \mathbf {U}_2^{\prime }\mathbf {Q}_p \varvec{\epsilon }=O_p(1)\), and \(\hat{\sigma }^2=O_p(1)\). Therefore, \(t_T(\lambda )_{\max }\) diverges to minus infinity as \(T \rightarrow \infty \). This also implies that the test statistic diverges to minus infinity under the alternative hypothesis as \(T \rightarrow \infty \).
Proof of Theorem 3
Part of (3a). Under \(H_0,\, \hat{\rho }\) is given by \(\hat{\rho }=(\mathbf {U}^{\prime } \mathbf {S}_p \mathbf {U})^{-1} \mathbf {U}^{\prime } \mathbf {S}_p \varvec{\epsilon }\), where \(\mathbf {S}_p=\mathbf {I}-\mathbf {N}_p(\mathbf {N}_p^{\prime }\mathbf {N}_p)^{-1}\mathbf {N}_p^{\prime }\) is a \(T \times T\) idempotent matrix and
Similar to (46), we can rewrite \(W_T^B(\lambda )\) as
Note that
Consider \((T^{-1} N_p N_p)^{-1}\). Since we obtain \(T^{-1}\sum _{t=1}^T \mathbf {1}\{\hat{u}_{t-1} \le \lambda _1 \} \Rightarrow \int _0^1 \mathbf {1} \{ W^*\le \tilde{\lambda }_1 \} \mathrm{d}r\), it can be shown that
where \(I_1=\mathbf {1} \{ W^*\le \tilde{\lambda }_1 \}\), \(I_2=\mathbf {1} \{ W^*> \tilde{\lambda }_2 \}\), and \( \varGamma =\lim _{T \rightarrow \infty } T^{-1}\sum E(\Delta \hat{u}_{t-p}^{p\prime } \Delta \hat{u}_{t-p}^p) \). Therefore, we have
We also have
and
It follows from Theorem 1, (62), (63), and (64) that
Next, consider \(\mathbf {U}_1^{\prime } \mathbf {S}_p \mathbf {U}_1\). It is easily seen that
Using (62) and (63), we obtain
Therefore, we have
Since similar analysis can be applied to \(\varvec{\epsilon }^{\prime } \mathbf {S}_p \mathbf {U}_2(\mathbf {U}_2^{\prime } S_p \mathbf {U}_2)^{-1} \mathbf {U}_2^{\prime } \mathbf {S}_p \varvec{\epsilon }\), it can also be shown that
and
Noting that \((\hat{\mu }_1-\mu _1)=O_p(T^{-1/2})\) and \((\hat{\mu }_2-\mu _2)=O_p(T^{-1/2})\), similar to (51), we obtain
We can deduce the required results from the asymptotic distribution of \(W_B^T(\lambda )\) and Assumption 2.
Under the alternative hypothesis, we can write \(W_T^B(\lambda )\) as
Since under the alternative hypothesis, we have \(T^{-1}\mathbf {U}^{\prime }\mathbf {S}_p \mathbf {U}=O_p(1)\), \(T^{-1/2}\mathbf {U}^{\prime } \mathbf {S}_p \varvec{\epsilon }=O_p(1)\), and \(\hat{\sigma }^2=O_p(1)\), it can be shown that
Therefore, \(W_T^B(\lambda )\) diverges to infinity as \(T \rightarrow \infty \). This also implies that the test statistic diverges to infinity as \(T \rightarrow \infty \).
Part of (3b) From the proof of part (2b) and (3a), under \(H_0\), the statistic \(t_T^B(\lambda )_{\max }\) can be written as
Using from (65) to (71), we have
We can deduce the required results from the asymptotic distribution of \(t_T^B(\lambda )_{\max }\) and Assumption 2. Next, we consider the properties of the test statistic under the alternative hypothesis. Under the alternative hypothesis, \(t_T^B(\lambda )_{\max }\) can be expressed as
Using \(T^{-1} \mathbf {U}_1^{\prime } \mathbf {S}_p \mathbf {U}_1=O_p(1),\, T^{-1/2} \mathbf {U}_1^{\prime }\mathbf {S}_p \varvec{\epsilon }=O_p(1),\, T^{-1} \mathbf {U}_2^{\prime } \mathbf {S}_p \mathbf {U}_2=O_p(1), T^{-1/2} \mathbf {U}_2^{\prime }\mathbf {S}_p \varvec{\epsilon }=O_p{1}\), and \(\hat{\sigma }^2=O_p(1)\), we obtain
which shows that \(t_T^B(\lambda )_{\max }\) diverges to minus infinity as \(T \rightarrow \infty \). This also implies that the test statistic diverges to minus infinity under the alternative hypothesis as \(T \rightarrow \infty \).
Appendix 2: Heteroskedasticity-robust test statistics
For (7), the heteroskedasticity-robust test statistic is given by
where \(W^{*}_T\) is the test statistics (6) corrected by the White estimator. \(W^{*}_T\) is given by
where \(\hat{\epsilon }\) is the residuals in regression (3). For the heteroskedasticity-robust test statistic of (10), we use White correction to calculate the standard error. The heteroskedasticity-robust versions for (15), (16), EG-\(t\), and ES-\(\phi \) are obtained similar to those above. See also Bec et al. (2010) for heteroskedasticity-robust unit root tests in smooth transition autoregressive (STAR) models.
Rights and permissions
About this article
Cite this article
Maki, D., Kitasaka, Si. Residual-based tests for cointegration with three-regime TAR adjustment. Empir Econ 48, 1013–1054 (2015). https://doi.org/10.1007/s00181-014-0822-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00181-014-0822-x