摘要
为探知BMI对各运动指标的影响,并对大学生体质变化趋势进行预测,跟踪采集某学院4年来20092名学生体测数据,首先引入拉以达法则剔除异常值,然后利用相关分析计量体能指标间的相关系数,用回归树模型对相关度进行了修正,最后基于均值GM(1,1)模型对于各体质指标进行了预测。结果显示男生BMI对引体向上影响最高,女生BMI对仰卧起坐影响最高,另模型中大多数指标预测平均相对误差小于1%,精度较高。
In order to find out the relationship between BMI and physical fitness test indicators, and to predict the trend of physical change of college students, to track the data of 20092 students in a college for 4 years, first introduce pauta criterion to eliminate the abnormal values, and then use the correlation analysis to measure the correlation coefficient between the physical indexes, and use the regression tree model to get the correlation degree. Finally, based on the mean GM （1,1） model, students＇ physical fitness indicators were predicted. The results showed that male BMI has the highest correlation with the upwards, female BMI has the highest correlation with sit ups, and the average relative error of most of the models was less than 1%, and the precision is higher. However, the function relationship between BMI and each index was not given because of the low goodness of fitting of the general measurement model.
作者
赵国瑞
郝世煜
田振明
ZHAO Guo-rui;HAO Shi-yu;TIAN Zhen-ming(Mathematical Model Laboratory,Guangzhou City Construction College,510925;School of Economics and Management,Guangzhou University of Chinese Medicine,510006 Guangzhou,China)
出处
《佳木斯大学学报（自然科学版）》
CAS
2018年第5期799-801,830,共4页
Journal of Jiamusi University：Natural Science Edition
基金
2018年广东省科技创新培育专项资金(攀登计划专项pdjhb0987)
广东省数学会课题([2016]15号)
