Intelligent Prediction of Suitable Physical Characteristics Toward Archery Performance Using Multivariate Techniques

Abstract

The present study aims to explore different physical characteristics towards a successful performance of archery and to predict the most vital attributes that contribute to the achievement of high archery scores. 32 archers drawn from different archery programmes participated in the study. Standard physical characteristics tests were conducted, and archers’ shooting scores of one end were recorded. Multivariate techniques of principal component analysis (PCA), hierarchical agglomerative cluster analysis (HACA) and discriminant analysis (DA) were used to achieve the aims of the study. The PCA after varimax rotation indicates two variables containing 12 and 2 varifacators (VF). The First VF revealing high positive loadings from weight (0.94), calf circumference (cc) (0.89), abdomen cc (0.97), hip cc (0.97), thigh cc (0.95), triceps thickness (0.76), biceps thickness (0.75), subscapular thickness (0.83), suprailiac thickness (0.85), abdominal thickness (0.85), front thigh thickness (0.76) and medial calf thickness (0.80) revealing that endomorphic body positively affect the performance of the sport. The second VF discloses high negative loadings from height (-0.88) and arm length (-0.90) describing that body height, and arm length negatively affects the performance of the sport. HACA classified the archers into three classes based on the PCA outputs namely; High-performance class, Medium-performance class and Low-performance class. Standard, backward and forward stepwise DA discriminate the classes from the 14 predicted variables with accuracies of 74.19%, 96.77% and 93.55% respectively. The findings from the current study can be helpful in mapping out potential athletes in archery based on their physical characteristics.

Keywords: Intelligent prediction, Archery, Physical characteristics, Multivariate techniques.