[第四章正交试验设计.ppt
文本预览下载声明
F检验 由偏差平方和的分解: 构成F统计量,进行F检验 当 时,拒绝原假设,检验显著。 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 偏差平方和的计算 列号 试验号 1 2 3 4 5 6 7 8 9 偏差平方和 1 A 1(800C) 1(800C) 1(800C) 2(850C) 2(850C) 2(850C) 3(900C) 3(900C) 3(900C) SA 2 B 1(90分) 2(120分) 3(150分) 1(90分) 2(120分) 3(150分) 1(90分) 2(120分) 3(150分) SB 3 C 1(甲) 2(乙) 3(丙) 2(乙) 3(丙) 1(甲) 3(丙) 1(甲) 2(乙) SC 4 1 2 3 3 1 2 2 3 1 Se 试验指标y 31 54 38 53 49 42 57 62 64 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 正交表每一列的自由度 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 其中 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. S1是由因子A的不同水平及随机误差引起,所以,SA=S1 同理, S2是由因子B的不同水平及随机误差引起, SB=S2 S3是由因子C的不同水平及随机误差引起, SC=S3 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. S4纯粹是由试验的随机误差引起,所以,S4=Se 不难证明 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 方差分析表 来源 A B C e T 平方和S 618 114 234 18 984 自由度f 2 2 2 2 8 均方和V 309 57 117 9 F比 34.33 6.33 13.00 显著性 * (*) Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. SPSS ANALYZE General Linear Model Univariate Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 方差分析表 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 进入到Options对话框,以计算各因子不同水平对应的均值 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. Estimated Marginal
显示全部