Abstract
Building on work of Lucas (Econom Theory 11:331–346, 1995a) we derive the limiting distribution for M-estimator based unit root tests in the ESTAR model. This yields that the LS based unit root tests by Kapetanios et al. (J Econom 112:359–379, 2003) and Rothe and Sibbertsen (Allg Stat Arch 90:439–456, 2006) are robustified in an outlier context. We also consider an LS based heteroscedasticity robust version (White’s) of one of our unit root tests as a “quick fix” solution to the problems of outliers. Finite sample properties of the tests are examined, and in the case of additive outliers it is shown that the LS based tests are grossly over-sized whereas the size of our tests is close to the nominal size. If an ESTAR model with innovation outliers is considered, significant power gains over the LS based tests are shown by using our robust tests.
Similar content being viewed by others
Notes
This can be explained by the fact that the influence function of M-estimators is unbounded in the space of explanatory variables, see Hampel et al. (1986, p. 313) for details.
The condition \(\theta \in \mathbb {R} _{+}\) is an identifying restriction.
It may also be noted that the result (a) in Theorem 2 of Rothe and Sibbertsen (2006, p. 444) can be obtained as a special case of our results by doing the corresponding substitutions for the parameters as described above, additionally noting that \(\mu _{\psi }=1\), into the right hand side of (i) in Theorem 1.
Evidently, the implied definition of the \(\tilde{t}_{KSS}\) statistic corresponds to the definition of the modified \(t\)-test statistic in Rothe and Sibbersten (2006, p. 445).
It is also seen the statistic \(\tilde{C}_{RS}\) merges as a special case of the \(\tilde{C}_{\psi ,\beta }\) statistic in (6) by substituting \(C_{\psi ,\beta }\) for \(T^{2}\hat{\beta }_{\epsilon }\) (the coefficient unit root test statistic in the LS case), and also apply the same replacements of consistent estimators as those previously described, additionally noticing that \(\mu _{\psi }=1\), to obtain the \(\tilde{t} _{\epsilon ,\beta }\) and \(\tilde{t}_{KSS}\) statistics.
The author is thankful for an anonymous referee pointing out this alternative solution to the nuisance parameter problem.
Corresponding table for the \(\tilde{C}_{\psi ,\beta }\) test is available upon request.
The author is thankful for an anonymous referee pointing out this.
For simplicity, it is assumed that the short-run dynamics enter in a linear way, and yields as such a correct approximation of the dynamics of the differences under the null hypothesis only.
All simulations are carried out in GAUSS 12 using the SIMD-oriented Fast Mersenne Twister 19937 random number generator.
Strictly speaking, the Huber \(\psi \)-function does not satisfy the differentiability condition of Assumption 2. However, the Huber \(\psi \) -function may be arbitrarily well approximated by the corresponding smooth Huber \(\psi \)-function, see Hampel et al. (2011), which satisfies the conditions of Assumption 2. Lastly, we expect the results in above theorem and corollaries to hold when the smoothness conditions on \(\psi \) are relaxed such that the Huber \(\psi \)-function (say) is allowed for, but then other techniques to prove our results are required.
This choice of \(c\) also implies that it is only (standardized) residuals \( \epsilon _{t}\) larger than (in absolute vale) \(1.345\) which receive less weights (the weighting function equals then \(1.345/|\epsilon _{t}|\)).
More specifically, denote the estimates of \(\hat{\beta }_{\psi }\) and \(\hat{ \sigma }_{\varepsilon }\) after the \(n\)th iteration by \(\hat{\beta }_{\psi }^{(n)}\) and \(\hat{\sigma }_{\varepsilon }^{(n)}\), respectively. Then we may compute \(\hat{\beta }_{\psi }^{(n+1)}\) as
$$\begin{aligned} \hat{\beta }_{\psi }^{(n+1)}=\frac{\sum _{t}\hat{\sigma }_{\varepsilon }^{(n)}\psi (\hat{e}_{t}^{(n)}/\hat{\sigma }_{\varepsilon }^{(n)})y_{t-1}^{3}\Delta y_{t}}{\sum _{t}\hat{\sigma }_{\varepsilon }^{(n)}\psi (\hat{e}_{t}^{(n)}/\hat{\sigma }_{\varepsilon }^{(n)})y_{t-1}^{6}} , \end{aligned}$$where \(\hat{e}_{t}^{(n)}=\Delta y_{t}-\hat{\beta }_{\psi }^{(n)}y_{t-1}^{3}\). We stopped this iteration when \(|\hat{\beta }_{\psi }^{(n+1)}-\hat{\beta } _{\psi }^{n}|<1\times 10^{-9}\). As a starting value for \(\hat{\beta }_{\psi }^{(0)}\) we considered the least median of squares estimator of Rousseeuw (1984). Finally, at each step we updated the scale (MAD) estimate (times \(1.483\)) using the residuals \(\{ \hat{e}_{t}^{(n)}\}\).
The DGP’s in this size study are inspired by the ones in Lucas (1995a).
The power results for these cases are available upon request from the author.
Percentiles for other values of \(\rho \) are available upon request.
References
Balke NS, Fomby TB (1994) Large shocks, small shocks, and economic fluctuations: outliers in macroeconomic time-series. J Appl Econom 9:181–200
Carstensen K (2003) The finite sample performance of robust unit root tests. Stat Pap 44:469–482
Davidson J (1994) Stochastic limit theory. Oxford University Press, Oxford
Demetrescu M (2010) On the Dickey–Fuller test with White standard errors. Stat Pap 51:11–25
Dickey DA, Fuller WA (1979) Distribution of the estimators for autoregressive time series. J Am Stat Assoc 74:427–431
Franses PH, Haldrup N (1994) The effects of additive outliers on tests for unit roots and cointegration. J Bus Econ Stat 12:471–478
Haldrup N (1994) Heteroscedasticity in non-stationary time series, some Monte Carlo evidence. Stat Pap 35:287–307
Haldrup N, Monta Nes A, Sans A (2011) Detection of additive outliers in seasonal time series. J Time Ser Econom 3:1–18
Hampel F, Henning C, Ronchetti E (2011) A smoothing principle for the Huber and other location M-estimators. Comput Stat Data Anal 55:324–337
Hampel FR, Ronchetti EM, Rousseeuw PJ, Stahel WA (1986) Robust statistics. The approach based on influence functions. Wiley, New york
Hansen BE (1992) Convergence to stochastic integrals for dependent heterogeneous processes. Econom Theory 8:489–501
Herce MH (1996) Asymptotic theory of LAD estimation in a unit root process with finite variance errors. Econom Theory 12:129–153
Kapetanios G, Shin Y, Snell A (2003) Testing for a unit root in the nonlinear STAR framework. J Econom 112:359–379
Kiliç R (2011) Testing for a unit root in a stationary ESTAR process. Econom Rev 30:274–302
Kruse R (2011) A new unit root test against ESTAR based on a class of modified statistics. Stat Pap 52:71–85
Lucas A (1995a) Unit root tests based on M estimators. Econom Theory 11:331–346
Lucas A (1995b) An outlier robust unit root test with an application to the extended Nelson–Plosser data. J Econom 66:153–173
Luukkonen R, Saikkonen P, Teräsvirta T (1988) Testing linearity against smooth transition autoregressive models. Biometrika 75:491–499
McDonald JB, Newey WK (1988) Partially adaptive estimation of regression models via the generalized t distribution. Econom Theory 4:428–457
Newey WK, West KD (1987) A simple, positive definite, heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica 55:277–301
Park JY, Phillips PCB (1988) Statistical inference in regressions with integrated processes: part 1. Econom Theory 4:468–497
Phillips PCB (1987) Time series regression with a unit root. Econometrica 55:277–301
Phillips PCB, Durlauf SN (1986) Multiple time series regression with integrated processes. Rev Econ Stud 53:473–496
Phillips PCB, Perron P (1988) Testing for unit root in time series regression. Biometrika 75:335–346
Rothe C, Sibbertsen P (2006) Phillips–Perron-type unit root tests in the nonlinear ESTAR framework. Allg Stat Arch 90:439–456
Rousseeuw PJ (1984) Least median of squares regression. J Am Stat Assoc 79:871–880
Said S, Dickey DA (1984) Testing for unit roots in autoregressive moving-average models of unknown order. Biometrika 71:599–607
Sandberg R (2009) Convergence to stochastic power integrals for dependent heterogeneous processes. Econom Theory 25:739–747
Schwert GW (1989) Tests for unit roots: a Monte Carlo investigation. J Bus Econ Stat 7:147–159
Shin DW, Sarkar S, Lee JH (1996) Unit root tests for time series with outliers. Stat Probab Lett 30:189–197
van Dijk D, Franses P, Lucas A (1999) Testing for smooth transition nonlinearity in the presence of outliers. J Bus Econ Stat 17:217–235
Vogelsang TJ (1998) Trend function hypothesis testing in the presence of serial correlation. Econometrica 66:123–148
White H (1980) A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica 48:817–838
Acknowledgments
The author is grateful two anonymous referees for very helpful suggestions and comments. The author is also thankful for financial support from Jan Wallander’s and Tom Hedelius’ Research Foundations, Grant No. P2012-0085:1
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1
The following Lemma will facilitate the proofs of Theorem 1.
Lemma
Given Assumptions 1 and 2, \(\delta =6\) , and \(x_{t}=x_{t-1}+\epsilon _{t}\). Then,
-
(a)
\(T^{-4}\sum _{t}x_{t-1}^{6}\psi (\epsilon _{t})^{2}\Rightarrow \sigma _{\psi }^{2}\bar{\sigma }_{\epsilon }^{6}\int b_{1}^{6},\)
-
(b)
\(T^{-4}\sum _{t}x_{t-1}^{6}\psi ^{\prime }(\epsilon _{t})\Rightarrow \mu _{\psi }\bar{\sigma }_{\epsilon }^{6}\int b_{1}^{6},\)
-
(c)
\(T^{-2}\sum _{t}x_{t-1}^{3}\psi (\epsilon _{t})\Rightarrow \bar{ \sigma }_{\psi }\bar{\sigma }_{\epsilon }^{3}\int b_{1}^{3}db_{2}+3/2\bar{ \sigma }_{\epsilon }^{2}(\bar{\sigma }_{\epsilon \psi }-\sigma _{\epsilon \psi })\int b_{1}^{2}.\)
Proof of Lemma
The proof of (a). Write
Here, \(T^{-4}\sigma _{\psi }^{2}\sum _{t}x_{t-1}^{6}\Rightarrow \sigma _{\psi }^{2}\bar{\sigma }_{\epsilon }^{6}\int b_{1}^{6}\) by the continuous mapping theorem, and because \(\{ \psi (\epsilon _{t})^{2}-\sigma _{\psi }^{2}\}\) under the present assumptions satisfies the conditions of Theorem 3.3 in Hansen (1992) it follows that
and thus \(T^{-4}\sum _{t}x_{t-1}^{6}\psi (\epsilon _{t})^{2}\Rightarrow \sigma _{\psi }^{2}\bar{\sigma }_{\epsilon }^{6}\int b_{1}^{6}\).
The proof of (b) is similar to the proof of (a) noticing that under the present assumptions \(\{ \psi ^{\prime }(\epsilon _{t})-\mu _{\psi }\}\) satisfies the conditions of Theorem 3.3 in Hansen (1992).
Next, the proof of (c). Along the lines in the proof of Theorem 1 in Sandberg (2009), from which the (sufficient) moment condition \(\delta =6\) is taken, it is straightforward to show that
where \(\sigma _{\epsilon \psi ,1}=\lim _{T\rightarrow \infty }T^{-1}\sum \nolimits _{i=1}^{T}\sum \nolimits _{j=i+1}^{\infty }\mathrm {E} \epsilon _{i}\psi (\epsilon _{j})\). The proof now follows by substituting for \(\int B_{1}^{3}dB_{2}=\bar{\sigma }_{\psi }\bar{\sigma }_{\epsilon }^{3}\int b_{1}^{3}db_{2}\), \(\sigma _{\epsilon \psi ,1}=(\bar{\sigma } _{\epsilon \psi }-\sigma _{\epsilon \psi })/2\), and \(\int B_{1}^{2}=\bar{ \sigma }_{\epsilon }^{2}\int b_{1}^{2}\) into (11). \(\square \)
Proof of Theorem 1
The proof of (i). A (appropriately scaled) Taylor-series approximation of ( 5) with respect to \(\hat{e}_{t}\) around \(\epsilon _{t}\) yields
where we have used that \(\hat{e}_{t}-\epsilon _{t}=-\hat{\beta }_{\psi }y_{t-1}^{3}\) holds under the null hypothesis, and \(R_{T}\) is a remainder term. Next, the triangle inequality, the first-order Lipschitz condition for \(\psi ^{\prime }(\epsilon _{t})\), and that \(\hat{e}_{t}-\epsilon _{t}=o_{p}(1)\) holds uniformly for all \(t\) implies
for some \(\kappa \in (0,\infty )\) (independent of \(\hat{e}_{t}\) and \( \epsilon _{t}\)). Hence, by (12) we can now write
Finally, apply parts (b) and (c) of the Lemma to (13) to complete the proof.
The proof of (ii). Write
Next, it is straightforward to show that \(T^{-4}\sum \nolimits _{t}y_{t-1}^{6} \psi ^{\prime }(\hat{e}_{t})=T^{-4}\sum \nolimits _{t}y_{t-1}^{6}\psi ^{\prime }(\epsilon _{t})+o_{p}(1)\) and \(T^{-4}\sum \nolimits _{t}y_{t-1}^{6}\psi (\hat{ e}_{t})^{2}=T^{-4}\sum \nolimits _{t}y_{t-1}^{6}\psi (\epsilon _{t})^{2}+o_{p}(1)\) since \(\hat{e}_{t}-\epsilon _{t}=o_{p}(1)\). It follows now by parts (a) and (b) in the Lemma that the denominator in the above expression for \(t_{\psi ,\beta }\) converges weakly to the square root of \( \sigma _{\psi }^{2}\mu _{\psi }^{-2}\bar{\sigma }_{\epsilon }^{-6}\left( \int b_{1}^{6}\right) ^{-1}\). Combine this result with the convergence result for the numerator in (13) in part (i) of the proof to yield the desirable result. \(\square \)
Proof of Corollary 3
The proof of (i). We shall first derive an equivalent expression for \(\int b_{1}^{3}db_{2}\). Thus, notice first that \(\left[ b_{1}(s),b_{2}(s)\right] ^{\prime }\sim BM(\tilde{\Omega })\) where
Next, a lower triangular decomposition of \(\tilde{\Omega }\) is given by
and we can write \(\left[ b_{1}(s),b_{2}(s)\right] ^{\prime }=\tilde{\Omega } ^{1/2}\left[ w_{1}(s),w_{2}(s)\right] ^{\prime }\) where \(w_{1}(s)\) and \( w_{2}(s)\) are two independent standard Brownian motions on \(C[0,1]\). Accordingly, substitute for the identities \(b_{1}=w_{1}\) and \(b_{2}=\rho w_{1}+\sqrt{1-\rho ^{2}}w_{2}\) into \(\int b_{1}^{3}db_{2}\) to obtain
where \(\eta \sim N(0,1)\) and is independent of \(w_{1}\). The last equality follows by using elementary results on Ito calculus and that \(\int w_{1}^{3}dw_{2}\) is a scale mixture of normal distributions. Thus, part (i) of Corollary 3 can also be written as
Lastly, noticing that \(\hat{\rho }\tilde{C}_{RS}\Rightarrow \rho \left( w_{1}^{4}(1)-6\int w_{1}^{2}\right) /(4\int w_{1}^{6})\) completes the proof.
The proof of (ii) is trivial. \(\square \)
Appendix 2
This appendix presents the 1, 5, and 10 % percentiles of the limiting distribution of \(\tilde{t}_{\psi ,\beta }\) in Corollary 2 as a function of \( \rho \).Footnote 17 As an estimate of \(\rho \) we may use \(\bar{s}_{\epsilon \psi }/(\bar{s}_{\epsilon } \bar{s}_{\psi })\) (i.e. the long-run sample correlation between \(\hat{e} _{t} \) and \(\psi (\hat{e}_{t})\)) (Table 6).
Rights and permissions
About this article
Cite this article
Sandberg, R. M-estimator based unit root tests in the ESTAR framework. Stat Papers 56, 1115–1135 (2015). https://doi.org/10.1007/s00362-014-0628-0
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-014-0628-0