Chemometrics is an interdisciplinary field that combines statistics and chemistry. From its earliest days, chemometrics has always been a practically oriented subdiscipline of analytical chemistry aimed at solving problems often overlooked by mainstream statisticians. An important example is solving multivariate calibration problems at reduced rank. The method of partial least-squares (PLS) was quickly recognized and embraced by the chemistry community long before many practitioners in statistics considered it worthy of a “second look.”
For many chemists, training in data analysis and statistics has been limited to the basic univariate topics covered in undergraduate analytical chemistry courses such as univariate hypothesis testing, for example, comparison of means. A few more details may have been covered in some senior-level courses on instrumental methods of analysis where topics such as univariate linear regression and prediction confidence intervals might be examined. In graduate school, perhaps a review of error propagation and analysis of variance (ANOVA) may have been encountered in a core course in analytical chemistry. These tools were typically introduced on a very practical level without a lot of the underlying theory. The chemistry curriculum simply did not allow sufficient time for more in-depth coverage. However, during the past two decades, chemometrics has emerged as an important subdiscipline, and the analytical chemistry curriculum has evolved at many universities to the point where a small amount of time is devoted to practical application-oriented introductions to some multivariate methods of data analysis.
Practical Guide to Chemometrics, Second Edition offers an accessible introduction to application-oriented multivariate methods of data analysis and procedures for solving a variety of problems using analytical chemistry and statistics. This second edition has been completely revised to feature new chapters on principal component analysis, self-modeling curve resolution, and multi-way analysis methods. It includes expanded material on normal distributions, sampling theory, signal processing, and digital filtering. It also discusses trends in chemometrics and identifies areas for future development. With summaries, solutions to problems, recommended reading, and references for each chapter, this text remains an essential resource for students, professionals, and researchers in the field.