Title
USING TIME SERIES TO STUDY EFFECT OF AIR TEMPERATURE ON BODY TEMPERATURE OF COWS IN PUERTO RICO
Abstract
Body temperature is an important measure for monitoring the health status of cows. The objective of this study is to (1) determine if a cow’s body temperature is related to ambient temperature, relative humidity, and/or temperature humidity index (THI); (2) look for signs of heat stress. The data are collected at five minute intervals during the summer months (December through February) in Puerto Rico. Regression analysis and a succession of time series analyses are conducted in time domains. Nonparametric spectral estimation and cross-spectra analysis are also performed in the frequency domain. A search for indications of heat stress is performed by characterizing the relationship between body temperature and environmental factors. Detailed approaches of regression with autocorrelated errors and transfer function model in time domain are presented, along with the comparison between two models.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Zeng, Y.; Parkhurst, A. M.; and Pantoja, J.
(2009).
"USING TIME SERIES TO STUDY EFFECT OF AIR TEMPERATURE ON BODY TEMPERATURE OF COWS IN PUERTO RICO,"
Conference on Applied Statistics in Agriculture.
https://doi.org/10.4148/2475-7772.1086
USING TIME SERIES TO STUDY EFFECT OF AIR TEMPERATURE ON BODY TEMPERATURE OF COWS IN PUERTO RICO
Body temperature is an important measure for monitoring the health status of cows. The objective of this study is to (1) determine if a cow’s body temperature is related to ambient temperature, relative humidity, and/or temperature humidity index (THI); (2) look for signs of heat stress. The data are collected at five minute intervals during the summer months (December through February) in Puerto Rico. Regression analysis and a succession of time series analyses are conducted in time domains. Nonparametric spectral estimation and cross-spectra analysis are also performed in the frequency domain. A search for indications of heat stress is performed by characterizing the relationship between body temperature and environmental factors. Detailed approaches of regression with autocorrelated errors and transfer function model in time domain are presented, along with the comparison between two models.
