Unit root test: Difference between revisions

Revision as of 22:04, 9 June 2016

Latest revision as of 12:15, 31 July 2025

(34 intermediate revisions by 16 users not shown)

Line 1:

{{Short description|Time series statistical test}}

In [[statistics]], a '''unit root test''' tests whether a [[time series]] variable is non-stationary using an [[autoregressive]] model.

In [[statistics]], a '''unit root test''' tests whether a [[time series]] variable is non-stationary and possesses a [[unit root]]. The null hypothesis is generally defined as the presence of a unit root and the alternative hypothesis is either [[Stationary process|stationarity]], [[Trend-stationary process|trend stationarity]] or explosive root depending on the test used.

== General approach ==

Line 5 ⟶ 6:

</math> can be written as,

:<math>y_t = D_t + z_t + \varepsilon_t </math>

where,

Line 18 ⟶ 19:

== Main tests ==

A commonly used test that is valid in large samples is the [[augmented Dickey–Fuller test]]. The optimal finite sample tests for a unit root in autoregressive models were developed by [[Denis Sargan]] and [[Alok Bhargava]] by extending the work by [[John von Neumann]], and [[James Durbin]] and [[Geoffrey Watson]]. In the observed time series cases, for example, Sargan-Bhargava statistics test the unit root null hypothesis in first order autoregressive models against one-sided alternatives, i.e., if the process is stationary or explosive under the alternative hypothesis.

Other popular tests include:

* [[augmented Dickey–Fuller test]]<ref>{{Cite journal | doi = 10.1080/01621459.1979.10482531| title = Distribution of the estimators for autoregressive time series with a unit root| year = 1979| last1 = Dickey | first1 = D. A. | last2 = Fuller | first2 = W. A. | journal = [[Journal of the American Statistical Association]] | volume = 74| issue = 366a| pages = 427–431}}</ref>

*: this is valid in large samples.

* [[Phillips–Perron test]]

* [[KPSS test]]
*: (in whichhere the null hypothesis is [[Trend -stationary process|trend stationarity]] rather than the presence of a [[Stationary process|stationarityunit root]]).

* [[ADF-GLS test]]

* [[Zivot–Andrews test]]

Unit root tests are closely linked to [[Autocorrelation|serial correlation]] tests. However, while all processes with a unit root will exhibit serial correlation, not all serially correlated time series will have a unit root. Popular serial correlation tests include:

* [[Breusch–Godfrey test]]

Line 31 ⟶ 33:

==Notes==

{{Notelist}}

{{Reflist}}

==References==

*{{cite book |last=Bierens, |first=H. J. (|year=2001). "|chapter=Unit roots", Ch. 29 in ''|title=A Companion to Econometric Theory'', |editor -first=B. |editor-last=Baltagi, |___location=Oxford: |publisher=[[Blackwell Publishers]], |pages=610–633. }} [http://econ.la.psu.edu/~hbierens/UNITROOT.PDF "2007 revision"] {{Webarchive|url=https://web.archive.org/web/20140617113943/http://econ.la.psu.edu/~hbierens/UNITROOT.PDF |date=2014-06-17 }}

* {{Cite journal | last1 = Bhargava | first1 = A. | authorlink = Alok Bhargava| title = On the theory of testing for unit roots in observed time series | journal = [[The Review of Economic Studies]] | volume = 53 | issue = 3 | pages = 369–384 | doi = 10.2307/2297634 | jstor = 2297634| year = 1986 | pmid = | pmc = }}

*{{cite book |last=Enders |first=Walter |title=Applied Econometric Time Series |publisher=[[John Wiley & Sons]] |year=2004 |edition=Second |pages=[https://archive.org/details/appliedeconometr00ende_0/page/170 170–175] |isbn=0-471-23065-0 |url-access=registration |url=https://archive.org/details/appliedeconometr00ende_0/page/170 }}

* Bierens, H.J. (2001). "Unit roots", Ch. 29 in ''A Companion to Econometric Theory'', editor B. Baltagi, Oxford: [[Blackwell Publishers]], 610–633. [http://econ.la.psu.edu/~hbierens/UNITROOT.PDF "2007 revision"]

*{{cite book |last=Maddala |first=G. S. |authorlink=G. S. Maddala |last2=Kim |first2=In-Moo |chapter=Issues in Unit Root Testing |title=Unit Roots, Cointegration, and Structural Change |url=https://archive.org/details/unitrootscointeg00madd |url-access=limited |___location=Cambridge |publisher=Cambridge University Press |year=1998 |isbn=0-521-58782-4 |pages=[https://archive.org/details/unitrootscointeg00madd/page/n116 98]–154 }}

* {{Cite journal | doi = 10.1080/01621459.1979.10482531| title = Distribution of the estimators for autoregressive time series with a unit root| year = 1979| last1 = Dickey | first1 = D. A. | last2 = Fuller | first2 = W. A. | journal = [[Journal of the American Statistical Association]] | volume = 74| issue = 366a| pages = 427–431}}

*{{cite book |last=Enders |first=Walter |title=Applied Econometric Time Series |publisher=[[John Wiley & Sons]] |year=2004 |edition=Second |pages=170–175 |isbn=0-471-23065-0 }}

*{{citation | first= K. | last= Patterson | title= Unit Root Tests in Time Series | volume= 1 | year= 2011 | publisher= [[Palgrave Macmillan]]}}.

*{{citation | first= K. | last= Patterson | title= Unit Root Tests in Time Series | volume= 2 | year= 2012 | publisher= [[Palgrave Macmillan]]}}.

* {{Cite journal | last1 = Sargan | first1 = J. D. | last2 = Bhargava | first2 = Alok | year = 1983 | title = Testing residuals from least squares regression for being generated by the Gaussian random walk | journal = [[Econometrica]] | volume = 51 | issue = 1 | pages = 153–174 | publisher = | jstor = 1912252 | doix = <!-- {{subst:#titleparts:{{subst:PAGENAME}}|0|2}} --> | url = | format = | accessdate = }}

[[Category:StatisticalTime series statistical tests]]

[[Category:Time series analysis]]