第七章方差分析(4)

2018-12-23 23:06

统计学 第七章 方差分析

① 将数据输入工作表中 ② 选择菜单“工具”—“数据分析”,打开“数据分析”对话框 ③ 选择其中的“方差分析:无重复双因素方差分析”,打开对话框,见图7.7 ④ 正确填写相关信息后,点“确定”,结果在F1到L36这个区域内显示,见图7.8 Excel解决方案

图7.7 “方差分析:可重复双因素方差分析” 分析工具对话框

图7.8 “方差分析:可重复双因素方差分析”结果截图

结论: ∵FA=6.73,F?=3.35,FA>F? 或 P-value=0.0043<0.05

∴拒绝原假设H01,即材料对输出电压的影响显著; ∵FB=47.25,F?=3.35,FB>F? 或 P-value=0.0000<0.05 ∴拒绝原假设H02,即环境温度对输出电压的影响显著; ∵FAB=6.55,F?=2.72,FAB>F? 或 P-value=0.0008<0.05 ∴拒绝原假设

H03,即材料与温度的交互对输出电压的影响显著。

第16页

统计学 第七章 方差分析

英文摘要与关键词

When the numerical measurements across three or more groups are continuous and certain assumptions are met, a methodology known as analysis of variance (ANOVA) may be employed to compare the means of the groups. The null hypothesis is not different in the population means. In a sense, the term ANOVA appears to be misnomer because the objective is to analyze differences among the group means. However, through an analysis of the variation in the data, both among and within variations, we are able to draw conclusions about possible differences in group means. “Within group” variation is considered an experimental error, while “among group” variation is attributable to treatment effects.

To perform an ANOVA test of equality of population means we subdivide the total variation in the measurements into two parts, that which is attributable to differences among the groups and that which is due to inherent variation within the groups. The total variation is usually represented by the sum of squares total (SST). The among-group variation usually called the sum of squares among groups (SSA), is measured by the sum of squared differences between the sample mean of each group and the overall mean, and then weighted by the sample size in each group. The within-group variation, usually called the sum of squares within groups, measures the difference between each observation and the mean of its own group and cumulates the squares of these differences over all groups. This variation only includes random error, so it called SSE for short.

If each of these sums of squares is divided by its associated degrees of freedom, we obtain three variance or mean square terms: MST, MSA and MSE. We compute the test statistic F as the ratio of the variance of MSA to MSE.

The results of an analysis of variance are usually displayed in an ANOVA summary table. Excel and many statistical software packages can give us that table. The entries in this table include the sources of variation, the degrees of freedom, the sums of squares, the mean squares, the calculated F statistic and the p-value. This allows us to make direct conclusions about the null hypothesis without referring to a table of critical values of the F distribution. If the p-value is less than the chosen level of significance, the null hypothesis is rejected.

We can also consider an experimental design model in which two factors are of interest as well as the effects of their interaction. Owing to the complexity of the calculations involved, particularly because the number of levels of each factor increases and the number of replications in each cell increases, we assume that in practice, Excel will be used when analyzing data obtained from factorial design models.

Key Words: Analysis of Variance (ANOVA), the sum of squares total, the among-group variation, the within-group variation, F statistic, two-way ANOVA, interaction effects

第17页

统计学 第七章 方差分析

习 题

一、单项选择题

1.某饮料生产企业研制了一种新型饮料,饮料有五种颜色。如果要考察颜色是否会影响销售量,则水平为( )。

A.2 B.3 C.4 D.5

2.下列指标中包含有系统性误差的是( )。 A.SSA B.SSE C. D.x

3.SST的自由度是( )。

A.r-1 B.n-r C.r-n D.n-1

4.单因素方差分析的备择假设应该是( )。 A.

xj???1??2??3?......??r B.?1,?2,?3,......,?r不全相等 ?,?,?,......,?r全不相等 D. 12C.123

5.如果要拒绝原假设,则下列式子( )必须成立。

?????3?......??r

A.

F?F? B.P-value<α C.F=1 D.P-value>α

6.对双因素方差分析(无交互作用),下列命题哪个是错的( )。

A.SST=SSA+SSB+SSE B.SSB的自由度是s-1 C.F临界值只有一个 D.必须对两个因素分别决策

7.如果要比较3种化肥(A、B两种新型化肥和传统化肥)施撒在三种类型(酸性、中性和碱性)的土地上对作物的产量情况有无差别,则往往考虑用( )方法。 A.单因素方差分析 B.三因素方差分析

C.无交互作用的双因素方差分析 D.有交互作用的双因素方差分析

二、多单项选择题

1.运用方差分析的前提条件是( )( )( )( )( )。

A.样本来自正态总体 B.各总体的均值相等 C.各总体的方差相等 D.各总体相互独立 E.样本必须是随机的

2.下列指标中包含有随机性误差的是( )( )( )( )( )。 A.SSA B.SSE C.SST D.MSA E.MSE

3. 用LSD方法进行多重比较,若t?/2(n?r)MSE(间没有显著差别( )( )( )( )( )。

A.x1?x4?3.2 B.x2?x4?5.7 C.x2?x5?2.3 D.x1?x3?1.8 E.x3?x5?4.1

4.对无交互作用的双因素方差分析表,下列命题哪个是对的( )( )( )( )( )。 A.SST=SSA+SSB+SSE B.SSB的自由度是n-1 C.可以计算三个F值 D.SST的自由度是n-1 E.F临界值可能是一个,也可能是两个

第18页

??????????11?)=2.5,则下列各式中哪些说明比较总体ninj统计学 第七章 方差分析

三、计算题

1.将24家生产产品大致相同的企业,按资金分为三类,每个公司的每100元销售收入的生产成本(单位:元)如下表。这些数据能否说明三类公司的市场生产成本有差异?(假定生产成本服从正态分布,且方差相同。α=0.05) 20-30 69 72 70 76 72 72 66 72 30-50 75 76 72 70 80 68 80 74 50以上 77 80 75 86 74 86 80 83

2.某SARS研究所对31名自愿者进行某项生理指标测试,结果如下表。这三类人的该项生理指标有差别吗?如果有差别,请进行多重比较分析。(假定该生理指标服从正态分布,且方差相同。α=0.05)

SARS患者 疑似者 非患者

3.为了解三种不同配比的饲料对仔猪影响的差异,对三种不同品种的猪各选三头进行试验,分别测得其三个月间体重增加量如下表所示。试分析不同饲料与不同品种对猪的生长有无显著差异。(假定其体重增加量服从正态分布,且方差相同。α=0.05) 体重增量 因 素 A 因 素 B 1.8 1.4 1.5 2.1 1.9 1.7 1.8 1.9 1.8 1.8 2.0 2.3 2.1 2.1 2.1 2.6 2.5 2.3 2.4 2.4 2.9 3.2 2.7 2.8 2.7 3.0 3.4 3.0 3.4 3.3 3.5 B1 30 31 27 B2 31 36 29 B3 32 32 28 A1 A2 A3

4.比较3种化肥(A、B两种新型化肥和传统化肥)施撒在三种类型(酸性、中性和碱性)的土地上对作物的产量情况有无差别,将每块土地分成6块小区,施用A、B两种新型化肥和传统化肥。收割后,测量各组作物的产量,得到的数据如下表。化肥、土地类型及其它们的交互作用对作物产量有影响吗?(假定作物产量服从正态分布,且方差相同。α=0.05) 化肥 种类 A B 土 地 中性 31, 32 36, 35 29,27 酸性 30, 35 31, 32 27, 25 碱性 32, 30 32, 30 28, 25 传统

第19页


第七章方差分析(4).doc 将本文的Word文档下载到电脑 下载失败或者文档不完整,请联系客服人员解决!

下一篇:罗寨小学学校工作计划上

相关阅读
本类排行
× 注册会员免费下载(下载后可以自由复制和排版)

马上注册会员

注:下载文档有可能“只有目录或者内容不全”等情况,请下载之前注意辨别,如果您已付费且无法下载或内容有问题,请联系我们协助你处理。
微信: QQ: