HETEROSKEDASTICITY-AUTOCORRELATION ROBUST TESTING USING BANDWIDTH EQUAL TO SAMPLE SIZE

By N. Kiefer, Timothy J. Vogelsang. Econometric Theory, 2002

180 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/ad07ed4e0e2c91c0f30bd3f48d6a5d1c51f802db

How researchers see this paper

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

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

Let's Fix It: Fixed-b Asymptotics versus Small-b Asymptotics in Heteroskedasticity and Autocorrelation Robust Inference (2010, unknown): “P fpF1(p; b) zg = Gp(z) +A(z)b+ o(b) (11)”
Testing for common deterministic trend slopes (2001, methodology): “The other two approaches are new and are based on extensions of the approach proposed by Kiefer and Vogelsang (2002a).”
HETEROSKEDASTICITY-AUTOCORRELATION ROBUST STANDARD ERRORS USING THE BARTLETT KERNEL WITHOUT TRUNCATION (2002, background): “Kiefer and Vogelsang (2002) analyze the case of a general kernel, k x , with continuous second derivative, k′′ x , and show that, ̂= ̂0+ T−1 ∑ j=1 k j/T ̂j + ̂ ′ j ⇒− ∫ 1”
A NEW ASYMPTOTIC THEORY FOR HETEROSKEDASTICITY-AUTOCORRELATION ROBUST TESTS (2005, unknown): “…forms of the densities are not available with the exception of t for the case of the Bartlett kernel with b 1 ~see Abadir and Paruolo, 2002; Kiefer and Vogelsang, 2002b!+ However, because the limiting distributions are simple functions of standard Brownian motions, critical values are…”
Spatial Heteroskedasticity and Autocorrelation Consistent Estimation of Covariance Matrix (2011, methodology): “Under the xed-b speci cation, the bandwidth is set proportional to the sample size and the associated test statistic converges to a non-standard limiting distribution (e.g. Kiefer and Vogelsang (2002, 2005)).”
Optimal Bandwidth Selection in Heteroskedasticity-Autocorrelation Robust Testing∗ (2005, unknown): “When δ 6= 0 and the second-order corrected critical values are used, we can use Theorem 1 to calculate the local asymptotic power, measured by P{ ¯̄̄ (δ +W (1))Ξ −1/2 b ¯̄̄ > zα,b}, as in the following corollary.”
Inference with dependent data using cluster covariance estimators (2011, methodology): “Tests based on the Kiefer and Vogelsang (2005) approximation behave similarly to tests based on the CCE, highlighting the similarity between the fixed-G approach for the CCE and the “fixed-b” approach for HAC estimators.”
Spectral Density Estimation and Robust Hypothesis Testing Using Steep Origin Kernels Without Truncation (2006, background): “As has been noted by Kiefer and Vogelsang (2002a, 2002b) and Jansson (2004) and as confirmed in the simulations reported here, such effects can have the advantage of better reflecting finite sample behavior in test statistics that employ LRV/HAC estimates leading to some improvement in test size.”
FIXED-b ASYMPTOTICS FOR SPATIALLY DEPENDENT ROBUST NONPARAMETRIC COVARIANCE MATRIX ESTIMATORS (2014, methodology): “In the time series case, Kiefer and Vogelsang (2002) show that, when bandwidth equals sample size,5 ̂T ⇒ [∫ 1 0 ∫ 1 0 ∂2 K (r,s) ∂r∂s Yp(r)Yp(s)drds ] ′, (3.1) where Yp is a p-vector of standard Brownian bridges.”
Low-Frequency Econometrics (2015, background): “…finitesample properties, a burgeoning literature studies inconsistent long-run variance estimators (see Kiefer, Vogelsang, and Bunzel (2000) and Kiefer and Vogelsang (2002, 2005), Jansson (2004), Phillips (2005), Sun, Phillips, and Jin (2008), Phillips, Sun and Jin (2006, 2007) and Gonçalves…”
Measuring Uncertainty about Long-Run Predictions (2016, methodology): “When the shape of the spectrum is unknown but parametrized, as in the bcd-model, 11This is analogous to the wider condence intervals that arise from the use of inconsistent HAC estimators; see Kiefer, Vogelsang, and Bunzel (2000) and Kiefer and Vogelsang (2002, 2005), and Müller (2014) for a review”
HAC Corrections for Strongly Autocorrelated Time Series (2014, background): “Interestingly, though, the original (Kiefer, Vogelsang, and Bunzel 2000; Kiefer and Vogelsang 2002b) suggestion of a Bartlett kernel estimator with bandwidth equal to the sample size can be written exactly as ω̂2KVB = T −1∑ l=1 κT,lY 2 l , (8) where 1/κT,l = 4T 2 sin(πl/(2T ))2 → π2l2 (see Appendix…”