Elsevier

Economics Letters

Volume 37, Issue 4, December 1991, Pages 399-403
Economics Letters

Moving average filters and unit roots

https://doi.org/10.1016/0165-1765(91)90077-XGet rights and content

Abstract

Moving average filtering a stationary AR(1) time series yields higher valued first order autocorrelations. Its implications for unit root testing in seasonally (un-) adjusted time series are evaluated theoretically, via simulations, as well as with an empirical example.

References (11)

There are more references available in the full text version of this article.

Cited by (16)

  • Analysing driver's decision in dilemma zone at signalized intersections under disordered traffic conditions

    2022, Transportation Research Part F: Traffic Psychology and Behaviour
    Citation Excerpt :

    Past studies also report that the smoothing technique could eliminate outliers and reduce the errors in the trajectory data (Kanagaraj et al., 2015; Venthuruthiyil & Chunchu, 2018), even when applied for data extracted using a semi-automated tool for disordered traffic conditions. Averaging is one form of smoothing (Punzo et al., 2011); hence a 5-point moving average technique is used to smoothen the extracted vehicular trajectories (Franses, 1991; Raju et al., 2017). After post-processing the data, the trajectory data is trimmed down to represent only the trajectories of vehicles after the onset and during the amber time.

  • A note on the application of the DF test to seasonal data

    2000, Statistics and Probability Letters
    Citation Excerpt :

    However, different implications arise for each case in the event of misspecification. In the case where the quarterly data is pre-filtered using (1+L+L2+L3) one obvious negative effect that may occur in empirical work concerns overdifferencing, which has implications on the power performance of the test procedure (see e.g., Franses, 1991, for a discussion of this problem). Alternatively, if the DF test is applied on yt one needs to ensure that the possible presence of seasonal unit roots is accounted for through the correct augmentation of the test regression; otherwise, incorrect inferences may be drawn due to the improper use of the limit distribution.

  • Periodic integration in quarterly UK macroeconomic variables

    1993, International Journal of Forecasting
  • The problem of annual inflation rate indicator

    2023, International Journal of Finance and Economics
View all citing articles on Scopus

Thanks are due to Eric Ghysels and Svend Hylleberg for helpful discussions.

View full text