KFUPM CHEM 642 Chemometrics Chemistry Department Spring 07/08 (072)
Textbook: Miller, J. C. and Miller, J. N., Statistics and Chemometrics for Analytical Chemistry, 5th ed. (Pearson, 2005).
Objectives Review statistical and chemometrics methods used in analytical chemistry Apply chemometrics for analytical chemistry cases
Learning Outcomes At the end of the course students will be able to 1. define accuracy and precision and calculate combinations of errors and confidence limits 2. differentiate between the application of parametric and non-parametric tests 3. understand the concept of significance testing, undertake simple significance tests and interpret significance probabilities 4. know how to construct appropriate calibration curves and extract quantitative information. A further important objective is to illustrate how such statistical information and calculations are obtained from common PC software. The examples class will be hands-on use of Excel.
References M. Meloun, J. Militky and M. Forina "Chemometrics for analytical chemistry", Ellis Horwoodseries in analytical chemistry, 1992. Mullins, E., Statistics for the Quality Control Laboratory, (RSC, 2003). Manly, B. F. J., Statistics for Environmental Science and Management, (Chapman & Hall, 2001).
Catalogue description CHEM 642 Chemometrics (3-0-3) Basic Statistics, Analysis of Variance (ANOVA), Computer Software (Mat Lab for Windows), Principles of Experimental Design, Factorial Designs and Analysis, Fractional Factorials, Response Surface Methodology, Second-order Designs, Application of the chemical Optimization by simplex. Prerequisite: CHEM 540 or equivalent
1 Topics # lectures 1. Errors in quantitative analysis 1 2. Statistics of Repeated Measurements 4 Mean and standard deviation The distribution of repeated measurements The sampling distribution of the mean Confidence limits of the mean for large samples Presentation of results Confidence limits of the geometric mean for a log-normal distribution Propagation of errors
3. Significance Tests 4 Comparison of an experimental mean with a known value Comparison of two experimental means Paired t-test One-sided and two-sided tests F-test for the comparison of standard deviations Outliers Analysis of variance Comparison of several means The arithmetic of ANOVA calculations The chi-squared test Testing for normality of distribution 4. The Quality of Analytical Measurements 4 Sampling Separation and estimation of variances using ANOVA Quality control methods Stewhart charts Establishing the process capability Average run length: cusum charts Proficiency testing schemes Collaborative trials Uncertainty Acceptable sampling 5. Calibration Methods in Instrumental Analysis 4 Calibration graphs in instrumental analysis The product-moment correlation coefficient The line of regression of y on x Errors in the slope and intercept of the regression line Calculation of a concentration and its random error Limits of detection The method of standard additions Use of regression lines for comparing analytical methods Weighted regression lines Intersection of two straight lines ANOVA and regression calculations Curve fitting Outliers in regression
2 6. Non-parametric and Robust Methods 4 The median: initial data analysis The sign test The Wald-Wolfowitz runs test The Wilcoxon signed rank test Simple tests for two independent samples Non-parametric tests for more than two samples Rank correlation Non-parametric regression methods Robust methods Robust regression methods The Kolmogorov test for goodness of fit 7. Experimental Design and Optimization 4 Randomization and blocking Two-way ANOVA Latin squares and other designs Interactions Factorial versus one-at-a-time design Factorial design and optimization Optimization: basic principles and univariate methods Optimization using the alternating variable search method The method of steepest ascent Simplex optimization Simulated annealing
8. Multivariate Analysis 5 Initial analysis Principal component analysis Cluster analysis Discriminate analysis K-nearest neighbor method Disjoint class modeling Multiple regression Principal component regression Multivariate regression Partial least squares regression Multivariate calibration Artificial neural networks
Examinations First major Exam: Monday, March 24, 2008 Second major exam: Monday, May 05, 2008 Final Exam: To be announce
Assessment The final grade will be based on the following: Assignments and presentations 30 Two major exams 40 Final exam 30 3