Quadratic model to estimate the doses causing the highest cholesterol concentration and the same cholesterol concentration as control group

Date
2003-01-01
Authors
Zjhang, Wuyan
Nissen, Steve
Beitz, Donald
Dixon, Philip
Dixon, Philip
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Authors
Research Projects
Organizational Units
Animal Science
Organizational Unit
Statistics
Organizational Unit
Journal Issue
Series
Department
Animal ScienceStatistics
Abstract

High plasma cholesterol (particularly high LDL-cholesterol) is a high risk factor for coronary heart disease (CHD), which causes a high CHD morbidity and mortality. Besides clinical drugs, more and more interest is focused on finding natural components in the diet that may have hypocholesterolemic effects. Plant sterols are natural components in human diets and found to have cholesterol-lowering effects in humans. Sheanut oil has a relatively high amolmt of plant sterols. Therefore, the two experiments were designed to investigate the hypocholesterolemic effect of sheanut oil in hamsters. The response was not monotonic. Low doses increased plasma cholesterol, but high doses decreased plasma cholesterol. Because there was partial dose repetition between the two experiments, the two were combined together to estimate the dose leading to the highest cholesterol concentration and the dose leading to the same cholesterol concentration as the control group. A quadratic model was selected to fit the combined data after appropriate transformation of exploratory and response variable. Nonparametric smoothing method was used to justify the quadratic model. The results of point estimation and confidence interval were compared by Delta, Fieller's and bootstrapping methods.

Comments

This proceeding is from Zhang, W., Nissen, S., Beitz, D., & Dixon, P. (2003). Quadratic model to estimate the doses causing the highest cholesterol concentration and the same cholesterol concentration as control group. 15th Annual Conference on Applied Statistics in Agriculture, Apr. 27-29, Manhattan, Kansas, Kansas State University. doi:10.4148/2475-7772.1176.

Description
Keywords
Citation
DOI