Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Estimation of nutrient requirements using broken-line regression analysis.

K R Robbins1, A M Saxton, L L Southern

  • 1Department of Animal Science, University of Tennessee, Knoxville, 37996-4588, USA. krobbins@utk.edu

Journal of Animal Science
|April 4, 2006
PubMed
Summary

This study compared broken-line regression models for estimating nutrient requirements using SAS software. The quadratic model with a random asymptote component in SAS NLMixed provided the best fit for dose-response data.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Short communication: Relationships among temperature-humidity index with rectal, udder surface, and vaginal temperatures in lactating dairy cows experiencing heat stress.

Journal of dairy science·2018
Same author

Bovine intramammary infection associated immunogenic surface proteins of Streptococcus uberis.

Microbial pathogenesis·2017
Same author

Relationships among temperament, acute and chronic cortisol and testosterone concentrations, and breeding soundness during performance testing of Angus bulls.

Theriogenology·2017
Same author

Hemlock Woolly Adelgid (Hemiptera: Adelgidae) Abundance and Hemlock Canopy Health Numerous Years After Imidacloprid Basal Drench Treatments: Implications for Management Programs.

Journal of economic entomology·2016
Same author

Genotype-dependent Metabolic Responses to Semi-Purified High-Sucrose High-Fat Diets in the TALLYHO/Jng vs. C57BL/6 Mouse during the Development of Obesity and Type 2 Diabetes.

Experimental and clinical endocrinology & diabetes : official journal, German Society of Endocrinology [and] German Diabetes Association·2016
Same author

Relationships among temperament, behavior, and growth during performance testing of bulls.

Journal of animal science·2015

Area of Science:

  • Animal Nutrition
  • Statistical Modeling
  • Biostatistics

Background:

  • Estimating nutrient requirements is crucial for animal growth and health.
  • Broken-line regression models are commonly used for analyzing nutrient dose-response data.
  • Various statistical approaches exist, necessitating a comparison of their efficacy.

Purpose of the Study:

  • To evaluate and compare different broken-line regression models for estimating nutrient requirements.
  • To assess the performance of SAS (Statistical Analysis System) procedures NLIN and NLMixed for this purpose.
  • To identify the most suitable model and statistical approach for analyzing nutrient dose-response data.

Main Methods:

  • Utilized SAS procedures NLIN and NLMixed to fit five distinct broken-line regression models.

Related Experiment Videos

  • Included simple two straight-line, one-breakpoint models and quadratic broken-line models.
  • Incorporated random components for plateau and/or slope in models fitted with NLMixed.
  • Employed swine isoleucine requirement data from Parr et al. (2003).

    • Main Results:

    • The quadratic broken-line model with a random component for the asymptote, fitted using SAS NLMixed, demonstrated the best statistical fit.
    • This model achieved a greater adjusted R-squared and the least log likelihood value.
    • SAS NLMixed generally provided superior fits compared to SAS NLIN for the tested models.

    Conclusions:

    • SAS NLMixed, specifically with a quadratic broken-line model incorporating a random asymptote component, is recommended for analyzing nutrient dose-response data.
    • The study provides valuable insights into model selection and statistical approaches for precise nutrient requirement estimation.
    • Detailed model descriptions and SAS code are provided for practical application.