Abstract
To capture location shifts in the context of model selection, we propose selecting significant step indicators from a saturating set added to the union of all of the candidate variables. The null retention frequency and approximate non-centrality of a selection test are derived using a ‘split-half’ analysis, the simplest specialization of a multiple-path block-search algorithm. Monte Carlo simulations, extended to sequential reduction, confirm the accuracy of nominal significance levels under the null and show retentions when location shifts occur, improving the non-null retention frequency compared to the corresponding impulse-indicator saturation (IIS)-based method and the lasso.
Highlights
Unmodelled location shifts can have pernicious effects on the constancy of models and on forecast performance
Hendry et al [3] derive the null distribution of indicator saturation (IIS) for independent, identically distributed (IID) data, and [4] generalize that analysis to dynamic regression models
We investigate the power of a step indicator to detect a known mean shift from λ1 6= 0 to λ1 = 0 at time 0 < T1 < T /2 in the data generation process (DGP): yt = μ + λ1 1{t≤T1 } + t where t ∼ IN 0, σ2 (6)
Summary
Unmodelled location shifts (changes in previous unconditional means of data) can have pernicious effects on the constancy of models and on forecast performance. Indicator saturation methods (such as IIS and SIS) are feasible because software, like Autometrics, can handle more candidate variables N than observations T during model selection using a combination of expanding and contracting multiple block searches, as described in [13], [14] (Chapter 19), and [15]. In this selection context, the null retention frequency of indicators is called the gauge by [16], akin to the size of a test denoting its (false) null rejection frequency, but taking into account that indicators that are insignificant on a pre-assigned criterion may be retained to offset what would otherwise be a significant misspecification test.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.