Robustness of joint regression analysis

Dulce G. Pereira; João Tiago Praça Nunes Mexia; Paulo Canas Rodrigues

Robustness of joint regression analysis

Dulce G. Pereira, João Tiago Praça Nunes Mexia, Paulo Canas Rodrigues

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

Abstract

Joint Regression Analysis is shown to be extremely robust to missing observations. Thus, using a series of "α-designs" of winter rye cultivars, it was shown that with up to 40% of missing observations the cultivars to be selected would be the same. In this study we considered missing observations incidences varying from 5% to 75% with 5% differences between them. For each incidence the positions of missing observations were randomly generated in triplicate.

Original language	English
Pages (from-to)	105-128
Number of pages	24
Journal	Biometrical Letters
Volume	44
Issue number	2
Publication status	Published - 1 Jan 2007

Keywords

Joint Regressions Analysis
Robustness
Missing observations
Linear regressions
L2 environmental indexes

Cite this

@article{07e3b9c7f9f14643b4b242bacee7f0f8,

title = "Robustness of joint regression analysis",

abstract = "Joint Regression Analysis is shown to be extremely robust to missing observations. Thus, using a series of {"}α-designs{"} of winter rye cultivars, it was shown that with up to 40% of missing observations the cultivars to be selected would be the same. In this study we considered missing observations incidences varying from 5% to 75% with 5% differences between them. For each incidence the positions of missing observations were randomly generated in triplicate.",

keywords = "Joint Regressions Analysis, Robustness, Missing observations, Linear regressions, L2 environmental indexes",

author = "Pereira, {Dulce G.} and Mexia, {Jo{\~a}o Tiago Pra{\c c}a Nunes} and Rodrigues, {Paulo Canas}",

year = "2007",

month = jan,

day = "1",

language = "English",

volume = "44",

pages = "105--128",

journal = "Biometrical Letters",

issn = "1896-3811",

publisher = "De Gruyter",

number = "2",

}

TY - JOUR

T1 - Robustness of joint regression analysis

AU - Pereira, Dulce G.

AU - Mexia, João Tiago Praça Nunes

AU - Rodrigues, Paulo Canas

PY - 2007/1/1

Y1 - 2007/1/1

N2 - Joint Regression Analysis is shown to be extremely robust to missing observations. Thus, using a series of "α-designs" of winter rye cultivars, it was shown that with up to 40% of missing observations the cultivars to be selected would be the same. In this study we considered missing observations incidences varying from 5% to 75% with 5% differences between them. For each incidence the positions of missing observations were randomly generated in triplicate.

AB - Joint Regression Analysis is shown to be extremely robust to missing observations. Thus, using a series of "α-designs" of winter rye cultivars, it was shown that with up to 40% of missing observations the cultivars to be selected would be the same. In this study we considered missing observations incidences varying from 5% to 75% with 5% differences between them. For each incidence the positions of missing observations were randomly generated in triplicate.

KW - Joint Regressions Analysis

KW - Robustness

KW - Missing observations

KW - Linear regressions

KW - L2 environmental indexes

M3 - Article

SN - 1896-3811

VL - 44

SP - 105

EP - 128

JO - Biometrical Letters

JF - Biometrical Letters

IS - 2

ER -

Robustness of joint regression analysis

Abstract

Keywords

Fingerprint

Cite this