Covariance Analysis of Censored Survival Data Using Log-Linear Analysis Techniques

Laird, Nan M.; Olivier, Donald C.

doi:10.2307/2287816

Cited by 224 publications

(144 citation statements)

References 0 publications

Supporting

Mentioning

140

Contrasting

Unclassified

Order By: Relevance

“…In the main hypothesis test we used extended log-linear models 29 to compare rates of low back injury in the intervention and control groups. Only a worker's first injury was counted.…”

Section: Discussionmentioning

confidence: 99%

A Controlled Trial of an Educational Program to Prevent Low Back Injuries

et al. 1997

View full text Add to dashboard Cite

“…In the main hypothesis test we used extended log-linear models 29 to compare rates of low back injury in the intervention and control groups. Only a worker's first injury was counted.…”

Section: Discussionmentioning

confidence: 99%

A Controlled Trial of an Educational Program to Prevent Low Back Injuries

et al. 1997

View full text Add to dashboard Cite

“…By construction of an appropriate design matrix estimation methods and software for binary response models can be used. Early references for the embedding into the framework of binary regression are Brown (1975) and Laird and Olivier (1981), see also Fahrmeir and Tutz (2001).…”

Section: Estimation With Censoringmentioning

confidence: 99%

“…The logistic model was later investigated by Thompson (1977) whereas Prentice and Gloeckler (1978) focussed on the grouped Cox model. The representation of hazard models as binary Bernoulli trials was propagated by Brown (1975) and Laird and Olivier (1981). Extensions to more flexible models with nonparametric components were proposed, for example, by Fahrmeir (1994), Fahrmeir and Knorr-Held (1997), Fahrmeir and Kneib (2011), Tutz and Pritscher (1996) and Kauermann et al (2005).…”

Section: Introductionmentioning

confidence: 99%

Variable selection in discrete survival models including heterogeneity

Groll

Tutz

2016

Lifetime Data Anal

View full text Add to dashboard Cite

Several variable selection procedures are available for continuous time-toevent data. However, if time is measured in a discrete way and therefore many ties occur models for continuous time are inadequate. We propose penalized likelihood methods that perform efficient variable selection in discrete survival modeling with explicit modeling of the heterogeneity in the population. The method is based on a combination of ridge and lasso type penalties that are tailored to the case of discrete survival. The performance is studied in simulation studies and an application to the birth of the first child.

show abstract

“…Age-adjusted rates for each year were calculated by an iterated proportional fitting process. 7 Incidence rates were expressed as rates per 1000 person-years at risk. Estimation of the change in incidence over time was by standard yearwise logistic regression analysis of the probability of having a stroke during a calendar year among subjects at risk at the start of the year.…”

Section: Methodsmentioning

confidence: 99%

Trends in stroke incidence and mortality in Hawaiian Japanese men.

Kagan

Popper

Reed

et al. 1994

Stroke

View full text Add to dashboard Cite

Vital statistics show a sharp decline in stroke mortality since the late 1960s. It is not clear whether this has been associated with a decline in stroke incidence. Since 1966 the Honolulu Heart Program has monitored the incidence and mortality of coronary heart disease and stroke in a target population of 11,136 men of Japanese ancestry living on Oahu. Trends were analyzed from January 1, 1969, through December 31, 1988. Of 7893 men aged 45 to 68 years and free of stroke at entry examination, 530 developed first episodes of stroke in the period 1969 to 1988 (389 cases of thromboembolic stroke, 124 cases of hemorrhagic stroke, and 17 cases of stroke of unknown type). Age-adjusted annual stroke incidence declined from 5.1 per 1000 person-years in 1969 to 1972 to 2.4 in 1985 to 1988. The incidences of thromboembolic stroke, hemorrhagic stroke, and total stroke decreased 3.5%, 4.2%, and 4.4% per year, respectively. The 1-month case-fatality rates for thromboembolic stroke decreased moderately; those for hemorrhagic stroke fell dramatically. These findings suggest that the decline in stroke mortality in the past two or three decades results from a decline in both incidence rates and early case-fatality rates in thromboembolic and hemorrhagic stroke and stroke of unknown type. The decreases may be related to changes in risk factors, such as the decline in blood pressure and the decrease in cigarette smoking, as well as improvements in diagnosis and treatment.

show abstract

Covariance Analysis of Censored Survival Data Using Log-Linear Analysis Techniques

Cited by 224 publications

References 0 publications

A Controlled Trial of an Educational Program to Prevent Low Back Injuries

A Controlled Trial of an Educational Program to Prevent Low Back Injuries

Variable selection in discrete survival models including heterogeneity

Trends in stroke incidence and mortality in Hawaiian Japanese men.

Contact Info

Product

Resources

About