A stepwise logistic regression analysis: An application toward poultry farm data in Johor

• Published in: International Journal of Engineering and Technology(UAE)
• Main Author: Jamil S.A.M.; Abdullah M.A.A.; Long K.S.; Jupri N.F.M.; Mamat M.
• Format: Article
• Language: English
• Published: Science Publishing Corporation Inc 2018
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082349915&partnerID=40&md5=6a01405aae47bb26746f7f74d7dd9320

The aims of this study are to fit a logistic regression model towards the fly problem in a farm and to identify the variables that are associated with the fly problem in a poultry farm. By using SPSS software, this study used 'FORWARD STEPWISE' and 'BACKWARD STEPWISE' methods to perform the analysis. Compared to linear regression analysis, logistic regression does not require rigorous assumptions to be met. This study used Likelihood Ratio test. Omnibus lest and Hosmer and Lemeshow test to validate and to test the fit of poultry farm data. Akaike Information Criterion (A1C) is calculated to observe the difference between the methods of stepwise used by SPSS software in this study. As a result, logistic regression is fit towards poultry farm data by a stepwise procedure. BACKWARD STEPWISE seems to be more suitable for conducting the stepwise method of analysis. Besides, variables that influence the problem of fly in a poultry arc the wasps, distance and number of flies. © 2018 Authors.