### Abstract

Original language | English |
---|---|

Place of Publication | Tilburg |

Publisher | Econometrics |

Number of pages | 10 |

Volume | 1997-105 |

Publication status | Published - 1997 |

### Publication series

Name | CentER Discussion Paper |
---|---|

Volume | 1997-105 |

### Fingerprint

### Keywords

- Jeffreys' prior
- multivariate regression model
- posterior existence
- scale mixture of normals

### Cite this

*Reference Priors for the General Location-Scale Model*. (CentER Discussion Paper; Vol. 1997-105). Tilburg: Econometrics.

}

**Reference Priors for the General Location-Scale Model.** / Fernández, C.; Steel, M.F.J.

Research output: Working paper › Discussion paper › Other research output

TY - UNPB

T1 - Reference Priors for the General Location-Scale Model

AU - Fernández, C.

AU - Steel, M.F.J.

N1 - Pagination: 10

PY - 1997

Y1 - 1997

N2 - The reference prior algorithm (Berger and Bernardo 1992) is applied to multivariate location-scale models with any regular sampling density, where we establish the irrelevance of the usual assumption of Normal sampling if our interest is in either the location or the scale. This result immediately extends to the linear regression model. On the other hand, an essentially arbitrary step in the reference prior algorithm, namely the choice of the nested sequence of sets in the parameter space is seen to play a role. Our results lend an additional motivation to the often used prior proportional to the inverse of the scale parameter, as it is found to be both the independence Jeffreys' prior and the reference prior under variation independence in the sequence of sets, for any choice of the sampling density. However, if our parameter of interest is not a one-to-one transformation of either location or scale, the choice of the sampling density is generally shown to intervene.

AB - The reference prior algorithm (Berger and Bernardo 1992) is applied to multivariate location-scale models with any regular sampling density, where we establish the irrelevance of the usual assumption of Normal sampling if our interest is in either the location or the scale. This result immediately extends to the linear regression model. On the other hand, an essentially arbitrary step in the reference prior algorithm, namely the choice of the nested sequence of sets in the parameter space is seen to play a role. Our results lend an additional motivation to the often used prior proportional to the inverse of the scale parameter, as it is found to be both the independence Jeffreys' prior and the reference prior under variation independence in the sequence of sets, for any choice of the sampling density. However, if our parameter of interest is not a one-to-one transformation of either location or scale, the choice of the sampling density is generally shown to intervene.

KW - Jeffreys' prior

KW - multivariate regression model

KW - posterior existence

KW - scale mixture of normals

M3 - Discussion paper

VL - 1997-105

T3 - CentER Discussion Paper

BT - Reference Priors for the General Location-Scale Model

PB - Econometrics

CY - Tilburg

ER -