An Automatic Test of Super Exogeneity

115 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Detecting Location Shifts during Model Selection by Step-Indicator Saturation (2015, background): “Then, ψ∗ λ1 in Equation (10) is √ T ∗ times the corresponding non-centrality for an individual impulse indicator in [5], so δ̂T1 will have considerable power when λ1 6= 0.”
MODEL DISCOVERY AND TRYGVE HAAVELMO’S LEGACY (2014, methodology): “The test for superexogeneity in Hendry and Santos (2010) uses IIS in an automatically created vector autoregressive model of {xt } to ascertain the occurrence and timing of location shifts like (δ2 −δ1) in-sample, then tests whether those enter the conditional model yt |xt , where their…”
Model selection when there are multiple breaks (2012, background): “Hendry and Santos (2010) present analytic ‘power’ calculations for IIS for a single outlier and a single level shift in a marginal process, assuming normality, but extending such results to multiple breaks is problematic as the detectability of any given break depends on that of all others, and at…”