Testing for parameter stability in quantile regression models
We propose a test for structural change of conditional distribution in dynamic regression models. The test is constructed based on time series regression quantile estimates and complements conventional parameter instability tests in least-square type regression models. Asymptotic distribution for our test under the null hypothesis is derived.
Year of publication: |
2008
|
---|---|
Authors: | Su, Liangjun ; Xiao, Zhijie |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 78.2008, 16, p. 2768-2775
|
Publisher: |
Elsevier |
Saved in:
Saved in favorites
Similar items by person
-
Adaptive nonparametric regression with conditional heteroskedasticity
Jin, Sainan, (2015)
-
A residual based test for the numm hypothesis of cointegration
Xiao, Zhijie, (1999)
-
Likelihood-based inference in trending time series with a root near unity
Xiao, Zhijie, (2001)
- More ...