Bayesian Tests of a Parameter Shift Under Heteroscedasticity: Weighted-T Vs. Double-T Approaches

Hiroki Tsurumi; Neil Sheflin

doi:10.1080/03610928408828736

Bayesian Tests of a Parameter Shift Under Heteroscedasticity: Weighted-T Vs. Double-T Approaches

Hiroki Tsurumi, Neil Sheflin

School of Arts and Sciences, Economics

Research output: Contribution to journal › Article › peer-review

6 Scopus citations

Abstract

A highest posterior density interval for examining shifts in individual regression coefficients between subsamples with different error variances is derived from a weighted-t posterior density. Compared to a double-t approach, the weighted-t approach is computationally more efficient and allows a direct comparison to a sampling theory approach. In a numerical example the weighted-t, double-t and confidence interval approaches are compared.

Original language	English (US)
Pages (from-to)	1003-1013
Number of pages	11
Journal	Communications in Statistics - Theory and Methods
Volume	13
Issue number	8
DOIs	https://doi.org/10.1080/03610928408828736
State	Published - Jan 1 1984

All Science Journal Classification (ASJC) codes

Statistics and Probability

Keywords

density intervals weighted-t
double-t densities
heteroscedastioity
highest posterior

Access to Document

10.1080/03610928408828736

Cite this

@article{53afb5a6a10d4e5f97f9df6000342481,

title = "Bayesian Tests of a Parameter Shift Under Heteroscedasticity: Weighted-T Vs. Double-T Approaches",

abstract = "A highest posterior density interval for examining shifts in individual regression coefficients between subsamples with different error variances is derived from a weighted-t posterior density. Compared to a double-t approach, the weighted-t approach is computationally more efficient and allows a direct comparison to a sampling theory approach. In a numerical example the weighted-t, double-t and confidence interval approaches are compared.",

keywords = "density intervals weighted-t, double-t densities, heteroscedastioity, highest posterior",

author = "Hiroki Tsurumi and Neil Sheflin",

year = "1984",

month = jan,

day = "1",

doi = "10.1080/03610928408828736",

language = "English (US)",

volume = "13",

pages = "1003--1013",

journal = "Communications in Statistics - Theory and Methods",

issn = "0361-0926",

publisher = "Taylor and Francis Ltd.",

number = "8",

}

TY - JOUR

T1 - Bayesian Tests of a Parameter Shift Under Heteroscedasticity

T2 - Weighted-T Vs. Double-T Approaches

AU - Tsurumi, Hiroki

AU - Sheflin, Neil

PY - 1984/1/1

Y1 - 1984/1/1

N2 - A highest posterior density interval for examining shifts in individual regression coefficients between subsamples with different error variances is derived from a weighted-t posterior density. Compared to a double-t approach, the weighted-t approach is computationally more efficient and allows a direct comparison to a sampling theory approach. In a numerical example the weighted-t, double-t and confidence interval approaches are compared.

AB - A highest posterior density interval for examining shifts in individual regression coefficients between subsamples with different error variances is derived from a weighted-t posterior density. Compared to a double-t approach, the weighted-t approach is computationally more efficient and allows a direct comparison to a sampling theory approach. In a numerical example the weighted-t, double-t and confidence interval approaches are compared.

KW - density intervals weighted-t

KW - double-t densities

KW - heteroscedastioity

KW - highest posterior

UR - http://www.scopus.com/inward/record.url?scp=84911519048&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84911519048&partnerID=8YFLogxK

U2 - 10.1080/03610928408828736

DO - 10.1080/03610928408828736

M3 - Article

AN - SCOPUS:84911519048

SN - 0361-0926

VL - 13

SP - 1003

EP - 1013

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

IS - 8

ER -

Bayesian Tests of a Parameter Shift Under Heteroscedasticity: Weighted-T Vs. Double-T Approaches

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this