COMPLETE CONVERGENCE OF ERROR VARIANCE ESITIMATES UNDER ϕ-MIXING ERROR IN LINEAR MODELS

1994 ◽  
Vol 14 (4) ◽  
pp. 417-425
Author(s):  
Liugen Xue
2017 ◽  
Vol 29 (2) ◽  
pp. 151-166 ◽  
Author(s):  
Hui-Ling Lin ◽  
Zhouping Li ◽  
Dongliang Wang ◽  
Yichuan Zhao

Author(s):  
Chris K. Bullough

A new procedure being developed in British Standards for the assessment of creep-rupture data is described, and evaluated with trial data sets of gas turbine blading materials. The procedure is applied in phases. An important development by statistical experts is a framework for the main assessment phase which uses maximum-likelihood fitting methods for the treatment of unfailed test points and error variance. The framework selects models from a standard suite (together with any other linear models supplied by the assessor) using statistical criteria, but also incorporates metallurgical judgement. The improved representation of the experimental data compared with previous fitting methods, and the associated statistical tests indicate that the new procedure can be used to derive rupture strength values for gas turbine materials with confidence.


2021 ◽  
pp. 140349482110313
Author(s):  
Jason A. Bennie ◽  
Guy Faulkner ◽  
Jordan J. Smith

Aims: This study aimed to describe the prevalence and socio-demographic and lifestyle-related correlates of muscle-strengthening activity (MSA; strength/resistance training, sit-ups/push-ups, etc.) among a large sample of European adolescents. Methods: Data were drawn from the European Health Interview Survey Wave 2 (2013–2014), including 8818 adolescents (15–17 years) from 28 European countries. Self-reported MSA was assessed using a previously validated survey item. Population-weighted prevalence ratios were calculated for (a) ‘none’ (0 days/week), (b) ‘insufficient MSA’ (1–2 days/week) or (c) ‘sufficient MSA’ (⩾3 days/week). Generalised linear models using Poisson regression with robust error variance were used to calculate the prevalence ratios for adolescents reporting sufficient MSA by socio-demographic/lifestyle characteristics and by European region. Results: Overall, 19.4% (95% confidence interval (CI) 18.3–20.7) reported sufficient (⩾3 days/week) MSA and 57.9% (95% CI 56.4–59.6) reported none. Females, adolescents from Southern and Eastern European regions, those not meeting the aerobic guideline and adolescents classified as overweight were significantly associated with a lower likelihood of reporting sufficient MSA, independent of other characteristics. Conclusions: The majority of European adolescents do not meet the MSA guidelines. Future large-scale MSA public-health interventions should target female and currently inactive adolescents, as well as those from Southern and Eastern European regions.


1998 ◽  
Vol 120 (3) ◽  
pp. 588-594
Author(s):  
C. K. Bullough

A new procedure being developed in British Standards for the assessment of creep-rupture data is described and evaluated with trial data sets of gas turbine blading materials. The procedure is applied in phases. An important development by statistical experts is a framework for the main assessment phase that uses maximum-likelihood fitting methods for the treatment of unfailed test points and error variance. The framework selects models from a standard suite (together with any other linear models supplied by the assessor) using statistical criteria, but also incorporates metallurgical judgement. The improved representation of the experimental data compared with previous fitting methods and the associated statistical tests indicate that the new procedure can be used to derive rupture strength values for gas turbine materials with confidence.


Sign in / Sign up

Export Citation Format

Share Document