随机数发生器在统计学案例式教学中的应用  

Application of Random Number Generator in the Statistics Case Teaching

作　　者：万海清 张家瑜 段雅丽 WAN Haiqing, ZHANG Jiayu, DUAN Yali(College of Life and Environmental Science, Hunan University of Arts and Scienc)

机构地区：[1]湖南文理学院生命与环境科学学院

出　　处：《农业网络信息》2018年第6期128-135,共8页Agriculture Network Information

基　　金：湖南文理学院教学改革研究项目"提升生命科学类本科生数据采集和分析能力的实践教学改革(编号:201542)";湖南文理学院研究性教学课程"果树栽培学总论"支持

摘　　要：为解决抽样分布理论教学与实践教学脱节的难题,保证应用统计学能够通过抽样误差分析大幅提升推断统计原理部分的教学效果。收集了一个含509个观测值的育种材料的数据资料,作为应用统计学案例式教学的母总体,使用Excel数据分析软件中的"随机数发生器"进行微型化的Monte-Carlo抽样研究,将正态分布教学的重心由描述性分布升级到抽样分布。并跨越抽样分布的大量样本模拟阶段,和三大统计分布一起,完成一系列抽样试验的容许范围占比分析,2016级生物科学、食品科学与工程、农学、动物科学等专业208人次上机实训的情况汇总统计,4种样本容量的抽样试验结果与理论概率的偏差经χ2适合性检验都不显著。To solve the problem of the disconnection between theory teaching and practice teaching in sampling distribution, and guarantee that application statistics can greatly raise the teaching effect of inferred statistical principle part by sampling error analysis, raw data on a breeding material containing 509 observations was collected as the original overall of applied statistics case teaching.The random number generator in the data analysis software Excel was used for the miniaturization of Monte-Carlo sampling study. The key point of normal distribution teaching was updated from descriptive distribution to sampling distribution. Beyond the large sample simulation of sampling distribution, the percent ratio analysis of allowable range of sampling trial was completed with three statistical distribution. In 2016-grade biology science, food science and engineering, agriculture, animal science et al., the actual training cases of 208 person-times were counted, the difference between the sampling test results and theoretical probability of the four sample sizes was not significant.

关 键 词：复置抽样 小概率原理 容许范围 推断统计 

分 类 号：G642[文化科学—高等教育学] O21-4[文化科学—教育学]