![]() ![]() In a final exam for a particular course at McMaster University there was an open-ended question. Use the 2% and 98% percentiles rather than the upper and lower hinge values. The whiskers on the plots are drawn at most 1.5 IQR distance units away from the box, however, if the whisker is to be drawn beyond the bound of the data vector, then it is redrawn at the edge of the data instead (i.e. The boxâs upper bound is at the 25th percentile, and the boxes lower bound is at the 75th percentile. Show outliers as dots, where an outlier is most commonly defined as any point 1.5 IQR distance units away from the box. Some variations for the box plot are possible: The median is also not balanced between the two quantiles for this box plot, when compared to the others. It is also the position with high variability, indicating that something about the saw blade at that position is not what it should be. In this figure we see how the thickness at position 1 is greater than at the other positions. The following box plot is a graphical summary of these numbers.Ī box plot is great for comparisons. Product development and product improvement Applications of Process Improvement using Data Analysis of designed experiments using PLS models Variability explained with each component A mathematical/statistical interpretation of PLS Advantages of the projection to latent structures (PLS) method ![]() Introduction to Projection to Latent Structures (PLS) Visualization latent variable models with linking and brushing Using indicator variables in a latent variable model Determining the number of components to use in the model with cross-validation Algorithms to calculate (build) PCA models Preprocessing the data before building a model Interpreting loadings and scores together More about the direction vectors (loadings) Extended topics related to designed experiments Blocking and confounding for disturbances Highly fractionated designs: beyond half-fractions Generators: to determine confounding due to blocking Generating the complementary half-fraction Example: analysis of systems with 4 factors Assessing significance of main effects and interactions Example: design and analysis of a three-factor experiment Analysis of a factorial design: interaction effects Analysis of a factorial design: main effects Changing one single variable at a time (COST) Experiments with a single variable at two levels Design and analysis of experiments in context Outliers: discrepancy, leverage, and influence of the observations More than one variable: multiple linear regression (MLR) Summary of steps to build and investigate a linear model Least squares models with a single x-variable The industrial practice of process monitoring Statistical tables for the normal- and t-distribution The normal distribution and checking for normality General summary: revealing complex data graphically ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |