Variable Selection in Prediction Equation by Ridge Regression Analysis for Body Composition Study

Autor: Minoru Yasuda, Norie Yamada, Shuichi Komiya, Kazutoshi Kikkawa
Rok vydání: 1995
Předmět:
Zdroj: Japan Journal of Human Growth and Development Research. 1995:63-66
ISSN: 1884-359X
1340-8682
Popis: The method of ridge regression analysis are most useful for analysis that are hindered by high correlations among the independent variables. In this study, prediction of the total body water (TBW) was attempted from standing height (HT), body weight (WT), and eight girth measurements of trunk and extremes (G0=chest girth, G1=neck, G2=abdominal, G3=hip, G4=upperarm, G5=forearm, G6=thigh, G7=lower leg) by this regression procedure. As the result, it was concluded that G2, G3, G4, Wt and G0 should be removed.
Databáze: OpenAIRE