A Weibull-based score test for heterogeneity.

A C Kimber

doi:10.1007/bf00128471

Home/Archive/PubMed

PubMed · 9384649

A Weibull-based score test for heterogeneity.

A C Kimber

View original source ↗Export metadata ↓

Abstract

The Weibull distribution is a natural starting point in the modelling of failure times in reliability, material strength data and many other applications that involve lifetime data. In recent years there has been a growing interest in modelling heterogeneity within this context. A natural approach is to consider a mixture, either discrete or continuous, of Weibull distributions. A judicious choice of mixing distribution yields a tractable and flexible generalization of the Weibull distribution. In this note a score test for detecting heterogeneity in this context is discussed and illustrated using some infant nutrition data.

Explore related subjects

Age Factors Biometry Breast Feeding Data Interpretation, Statistical Female Food Hypersensitivity Humans Infant Infant Nutritional Physiological Phenomena Life Tables Likelihood Functions Models, Statistical Journal Article Research Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

A C Kimber. 1996. A Weibull-based score test for heterogeneity.. https://doi.org/10.1007/bf00128471

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2007

Excess of cancers in Europe: a study of eleven major cancers amenable to lifestyle change.

Worldwide an estimated 11 million cancer cases were diagnosed in 2002, one quarter being in Europe. We estimated the potential in avoidable numbers and proportions of 11 cancers amenable to prevention (cancers of the oral cavity, oesophagus, stomach, colorectal, pancreas, laryngeal, lung, female breast, endometrium, kidney and bladder) in 28 European countries. We assumed that the aggregated rate of 3 countries with lowest incidence to be attainable throughout Europe. The difference between the age- and gender-specific national cancer incidence rates and the lowest rate observed in 2002 was determined and defined as "avoidable." Of the 1.4 million adult cases of selected cancers and countries within our study, 363,000 (59%) cancers in males and 326,000 (45%) cancers in females were hypothetically avoidable. Among men, the proportion was largest in Hungary (77%) and among women, in Belgium (54%). Assuming that differences in cancer incidence are not attributable to genetic susceptibility or diagnostic activity, about 50% of all cases of these 11 cancers could be potentially avoided, especially by decreased smoking among men. Interventions directed at reducing smoking, obesity and alcohol use as well as increasing physical activity and fruit and vegetable intake are necessary to attain lower incidence rates. It is important to recognize that the actual preventable cancer by eliminating currently known risk factors is somewhat less than we have estimated.

Age Factors↗

PubMed2007

Choice of time scale and its effect on significance of predictors in longitudinal studies.

Time-to-event regression is a frequent tool in biomedical research. In clinical trials this time is usually measured from the beginning of the study. The same approach is often adopted in the analysis of longitudinal observational studies. However, in recent years there has appeared literature making a case for the use of the date of birth as a starting point, and thus utilize age as the time-to-event. In this paper, we explore different types of age-scale models and compare them with time-on-study models in terms of the estimated regression coefficients they produce. We consider six proportional hazards regression models that differ in the choice of time scale and in the method of adjusting for the years before the study. By considering the estimating equations of these models as well as numerical simulations we conclude that correct adjustment for the age at entry is crucial in reducing bias of the estimated coefficients. The unadjusted age-scale model is inferior to any of the five other models considered, regardless of their choice of time scale. Additionally, if adjustment for age at entry is made, our analyses show very little to suggest that there exists any practically meaningful difference in the estimated regression coefficients depending on the choice of time scale. These findings are supported by four practical examples from the Framingham Heart Study.

Age Factors↗

PubMed2007

Reproductive risk factors for incident bladder cancer: Iowa Women's Health Study.

We studied the association between reproductive factors and bladder cancer incidence in a prospective cohort study of 37,459 Iowa women aged 55-69 years and initially free from cancer in 1986. Women reported reproductive history and were followed prospectively through 2003. After adjusting for age and smoking, there was an inverse association between age at menopause and incident bladder cancer (n = 192). Compared with menopause at age > or =48, the hazard ratio (HR) of bladder cancer was 1.32 (95% CI; 0.90-1.94) for menopause at 43-47, and 1.60 (95% CI; 1.06-2.39) for < or =42 (p-trend = 0.02). The associations were similar for ages at natural and surgical menopause. In addition, women with a history of bilateral oophorectomy had an increased risk of bladder cancer compared with those who did not undergo bilateral oophorectomy: HR = 1.58 (95% CI; 1.12, 2.22). Finally, there was an indication of a positive association between bladder cancer and shorter lifetime years of ovulation (p-trend = 0.09). There were no associations between incident bladder cancer and age at first birth, number of births, age at menarche, use of hormone replacement therapy or any other reproductive characteristics. This study provides evidence that increased risk of bladder cancer is associated with earlier age at menopause in postmenopausal women.

Age Factors↗