Chemometrics

Last updated

Chemometrics is the science of extracting information from chemical systems by data-driven means. Chemometrics is inherently interdisciplinary, using methods frequently employed in core data-analytic disciplines such as multivariate statistics, applied mathematics, and computer science, in order to address problems in chemistry, biochemistry, medicine, biology and chemical engineering. In this way, it mirrors other interdisciplinary fields, such as psychometrics and econometrics.

Contents

Background

Chemometrics is applied to solve both descriptive and predictive problems in experimental natural sciences, especially in chemistry. In descriptive applications, properties of chemical systems are modeled with the intent of learning the underlying relationships and structure of the system (i.e., model understanding and identification). In predictive applications, properties of chemical systems are modeled with the intent of predicting new properties or behavior of interest. In both cases, the datasets can be small but are often large and complex, involving hundreds to thousands of variables, and hundreds to thousands of cases or observations.

Chemometric techniques are particularly heavily used in analytical chemistry and metabolomics, and the development of improved chemometric methods of analysis also continues to advance the state of the art in analytical instrumentation and methodology. It is an application-driven discipline, and thus while the standard chemometric methodologies are very widely used industrially, academic groups are dedicated to the continued development of chemometric theory, method and application development.

Origins

Although one could argue that even the earliest analytical experiments in chemistry involved a form of chemometrics, the field is generally recognized to have emerged in the 1970s as computers became increasingly exploited for scientific investigation. The term 'chemometrics' was coined by Svante Wold in a 1971 grant application, [1] and the International Chemometrics Society was formed shortly thereafter by Svante Wold and Bruce Kowalski, two pioneers in the field. Wold was a professor of organic chemistry at Umeå University, Sweden, and Kowalski was a professor of analytical chemistry at University of Washington, Seattle. [2]

Many early applications involved multivariate classification, numerous quantitative predictive applications followed, and by the late 1970s and early 1980s a wide variety of data- and computer-driven chemical analyses were occurring.

Multivariate analysis was a critical facet even in the earliest applications of chemometrics. Data from infrared and UV/visible spectroscopy are often counted in thousands of measurements per sample. Mass spectrometry, nuclear magnetic resonance, atomic emission/absorption and chromatography experiments are also all by nature highly multivariate. The structure of these data was found to be conducive to using techniques such as principal components analysis (PCA), partial least-squares (PLS), orthogonal partial least-squares (OPLS), and two-way orthogonal partial least squares (O2PLS). [3] This is primarily because, while the datasets may be highly multivariate there is strong and often linear low-rank structure present. PCA and PLS have been shown over time very effective at empirically modeling the more chemically interesting low-rank structure, exploiting the interrelationships or 'latent variables' in the data, and providing alternative compact coordinate systems for further numerical analysis such as regression, clustering, and pattern recognition. Partial least squares in particular was heavily used in chemometric applications for many years before it began to find regular use in other fields.

Through the 1980s three dedicated journals appeared in the field: Journal of Chemometrics , Chemometrics and Intelligent Laboratory Systems , and Journal of Chemical Information and Modeling . These journals continue to cover both fundamental and methodological research in chemometrics. At present, most routine applications of existing chemometric methods are commonly published in application-oriented journals (e.g., Applied Spectroscopy, Analytical Chemistry , Analytica Chimica Acta, Talanta ). Several important books/monographs on chemometrics were also first published in the 1980s, including the first edition of Malinowski's Factor Analysis in Chemistry, [4] Sharaf, Illman and Kowalski's Chemometrics, [5] Massart et al. Chemometrics: a textbook, [6] and Multivariate Calibration by Martens and Naes. [7]

Some large chemometric application areas have gone on to represent new domains, such as molecular modeling and QSAR, cheminformatics, the '-omics' fields of genomics, proteomics, metabonomics and metabolomics, process modeling and process analytical technology.

An account of the early history of chemometrics was published as a series of interviews by Geladi and Esbensen. [8] [9]

Techniques

Multivariate calibration

Many chemical problems and applications of chemometrics involve calibration. The objective is to develop models which can be used to predict properties of interest based on measured properties of the chemical system, such as pressure, flow, temperature, infrared, Raman, [10] NMR spectra and mass spectra. Examples include the development of multivariate models relating 1) multi-wavelength spectral response to analyte concentration, 2) molecular descriptors to biological activity, 3) multivariate process conditions/states to final product attributes. The process requires a calibration or training data set, which includes reference values for the properties of interest for prediction, and the measured attributes believed to correspond to these properties. For case 1), for example, one can assemble data from a number of samples, including concentrations for an analyte of interest for each sample (the reference) and the corresponding infrared spectrum of that sample. Multivariate calibration techniques such as partial-least squares regression, or principal component regression (and near countless other methods) are then used to construct a mathematical model that relates the multivariate response (spectrum) to the concentration of the analyte of interest, and such a model can be used to efficiently predict the concentrations of new samples.

Techniques in multivariate calibration are often broadly categorized as classical or inverse methods. [7] [11] The principal difference between these approaches is that in classical calibration the models are solved such that they are optimal in describing the measured analytical responses (e.g., spectra) and can therefore be considered optimal descriptors, whereas in inverse methods the models are solved to be optimal in predicting the properties of interest (e.g., concentrations, optimal predictors). [12] Inverse methods usually require less physical knowledge of the chemical system, and at least in theory provide superior predictions in the mean-squared error sense, [13] [14] [15] and hence inverse approaches tend to be more frequently applied in contemporary multivariate calibration.

The main advantages of the use of multivariate calibration techniques is that fast, cheap, or non-destructive analytical measurements (such as optical spectroscopy) can be used to estimate sample properties which would otherwise require time-consuming, expensive or destructive testing (such as LC-MS). Equally important is that multivariate calibration allows for accurate quantitative analysis in the presence of heavy interference by other analytes. The selectivity of the analytical method is provided as much by the mathematical calibration, as the analytical measurement modalities. For example, near-infrared spectra, which are extremely broad and non-selective compared to other analytical techniques (such as infrared or Raman spectra), can often be used successfully in conjunction with carefully developed multivariate calibration methods to predict concentrations of analytes in very complex matrices.

Classification, pattern recognition, clustering

Supervised multivariate classification techniques are closely related to multivariate calibration techniques in that a calibration or training set is used to develop a mathematical model capable of classifying future samples. The techniques employed in chemometrics are similar to those used in other fields – multivariate discriminant analysis, logistic regression, neural networks, regression/classification trees. The use of rank reduction techniques in conjunction with these conventional classification methods is routine in chemometrics, for example discriminant analysis on principal components or partial least squares scores.

A family of techniques, referred to as class-modelling or one-class classifiers, are able to build models for an individual class of interest. [16] Such methods are particularly useful in the case of quality control and authenticity verification of products.

Unsupervised classification (also termed cluster analysis) is also commonly used to discover patterns in complex data sets, and again many of the core techniques used in chemometrics are common to other fields such as machine learning and statistical learning.

Multivariate curve resolution

In chemometric parlance, multivariate curve resolution seeks to deconstruct data sets with limited or absent reference information and system knowledge. Some of the earliest work on these techniques was done by Lawton and Sylvestre in the early 1970s. [17] [18] These approaches are also called self-modeling mixture analysis, blind source/signal separation, and spectral unmixing. For example, from a data set comprising fluorescence spectra from a series of samples each containing multiple fluorophores, multivariate curve resolution methods can be used to extract the fluorescence spectra of the individual fluorophores, along with their relative concentrations in each of the samples, essentially unmixing the total fluorescence spectrum into the contributions from the individual components. The problem is usually ill-determined due to rotational ambiguity (many possible solutions can equivalently represent the measured data), so the application of additional constraints is common, such as non-negativity, unimodality, or known interrelationships between the individual components (e.g., kinetic or mass-balance constraints). [19] [20]

Other techniques

Experimental design remains a core area of study in chemometrics and several monographs are specifically devoted to experimental design in chemical applications. [21] [22] Sound principles of experimental design have been widely adopted within the chemometrics community, although many complex experiments are purely observational, and there can be little control over the properties and interrelationships of the samples and sample properties.

Signal processing is also a critical component of almost all chemometric applications, particularly the use of signal pretreatments to condition data prior to calibration or classification. The techniques employed commonly in chemometrics are often closely related to those used in related fields. [23] Signal pre-processing may affect the way in which outcomes of the final data processing can be interpreted. [24]

Performance characterization, and figures of merit Like most arenas in the physical sciences, chemometrics is quantitatively oriented, so considerable emphasis is placed on performance characterization, model selection, verification & validation, and figures of merit. The performance of quantitative models is usually specified by root mean squared error in predicting the attribute of interest, and the performance of classifiers as a true-positive rate/false-positive rate pairs (or a full ROC curve). A recent report by Olivieri et al. provides a comprehensive overview of figures of merit and uncertainty estimation in multivariate calibration, including multivariate definitions of selectivity, sensitivity, SNR and prediction interval estimation. [25] Chemometric model selection usually involves the use of tools such as resampling (including bootstrap, permutation, cross-validation).

Multivariate statistical process control (MSPC), modeling and optimization accounts for a substantial amount of historical chemometric development. [26] [27] [28] Spectroscopy has been used successfully for online monitoring of manufacturing processes for 30–40 years, and this process data is highly amenable to chemometric modeling. Specifically in terms of MSPC, multiway modeling of batch and continuous processes is increasingly common in industry and remains an active area of research in chemometrics and chemical engineering. Process analytical chemistry as it was originally termed, [29] or the newer term process analytical technology continues to draw heavily on chemometric methods and MSPC.

Multiway methods are heavily used in chemometric applications. [30] [31] These are higher-order extensions of more widely used methods. For example, while the analysis of a table (matrix, or second-order array) of data is routine in several fields, multiway methods are applied to data sets that involve 3rd, 4th, or higher-orders. Data of this type is very common in chemistry, for example a liquid-chromatography / mass spectrometry (LC-MS) system generates a large matrix of data (elution time versus m/z) for each sample analyzed. The data across multiple samples thus comprises a data cube. Batch process modeling involves data sets that have time vs. process variables vs. batch number. The multiway mathematical methods applied to these sorts of problems include PARAFAC, trilinear decomposition, and multiway PLS and PCA.

Related Research Articles

<span class="mw-page-title-main">Analytical chemistry</span> Study of the separation, identification, and quantification of matter

Analytical chemistry studies and uses instruments and methods to separate, identify, and quantify matter. In practice, separation, identification or quantification may constitute the entire analysis or be combined with another method. Separation isolates analytes. Qualitative analysis identifies analytes, while quantitative analysis determines the numerical amount or concentration.

<span class="mw-page-title-main">Electron ionization</span> Ionization technique

Electron ionization is an ionization method in which energetic electrons interact with solid or gas phase atoms or molecules to produce ions. EI was one of the first ionization techniques developed for mass spectrometry. However, this method is still a popular ionization technique. This technique is considered a hard ionization method, since it uses highly energetic electrons to produce ions. This leads to extensive fragmentation, which can be helpful for structure determination of unknown compounds. EI is the most useful for organic compounds which have a molecular weight below 600. Also, several other thermally stable and volatile compounds in solid, liquid and gas states can be detected with the use of this technique when coupled with various separation methods.

<span class="mw-page-title-main">Calibration curve</span> Method for determining the concentration of a substance in an unknown sample

In analytical chemistry, a calibration curve, also known as a standard curve, is a general method for determining the concentration of a substance in an unknown sample by comparing the unknown to a set of standard samples of known concentration. A calibration curve is one approach to the problem of instrument calibration; other standard approaches may mix the standard into the unknown, giving an internal standard. The calibration curve is a plot of how the instrumental response, the so-called analytical signal, changes with the concentration of the analyte.

Analytical technique is a method used to determine a chemical or physical property of a chemical substance, chemical element, or mixture. There is a wide variety of techniques used for analysis, from simple weighing to advanced techniques using highly specialized instrumentation.

<span class="mw-page-title-main">Near-infrared spectroscopy</span> Analytical method

Near-infrared spectroscopy (NIRS) is a spectroscopic method that uses the near-infrared region of the electromagnetic spectrum. Typical applications include medical and physiological diagnostics and research including blood sugar, pulse oximetry, functional neuroimaging, sports medicine, elite sports training, ergonomics, rehabilitation, neonatal research, brain computer interface, urology, and neurology. There are also applications in other areas as well such as pharmaceutical, food and agrochemical quality control, atmospheric chemistry, combustion research and knowledge.

<span class="mw-page-title-main">Metabolomics</span> Scientific study of chemical processes involving metabolites

Metabolomics is the scientific study of chemical processes involving metabolites, the small molecule substrates, intermediates, and products of cell metabolism. Specifically, metabolomics is the "systematic study of the unique chemical fingerprints that specific cellular processes leave behind", the study of their small-molecule metabolite profiles. The metabolome represents the complete set of metabolites in a biological cell, tissue, organ, or organism, which are the end products of cellular processes. Messenger RNA (mRNA), gene expression data, and proteomic analyses reveal the set of gene products being produced in the cell, data that represents one aspect of cellular function. Conversely, metabolic profiling can give an instantaneous snapshot of the physiology of that cell, and thus, metabolomics provides a direct "functional readout of the physiological state" of an organism. There are indeed quantifiable correlations between the metabolome and the other cellular ensembles, which can be used to predict metabolite abundances in biological samples from, for example mRNA abundances. One of the ultimate challenges of systems biology is to integrate metabolomics with all other -omics information to provide a better understanding of cellular biology.

<span class="mw-page-title-main">Cyclic voltammetry</span> Method of analyzing electrochemical reactions

In electrochemistry, cyclic voltammetry (CV) is a type of potentiodynamic measurement. In a cyclic voltammetry experiment, the working electrode potential is ramped linearly versus time. Unlike in linear sweep voltammetry, after the set potential is reached in a CV experiment, the working electrode's potential is ramped in the opposite direction to return to the initial potential. These cycles of ramps in potential may be repeated as many times as needed. The current at the working electrode is plotted versus the applied voltage to give the cyclic voltammogram trace. Cyclic voltammetry is generally used to study the electrochemical properties of an analyte in solution or of a molecule that is adsorbed onto the electrode.

<span class="mw-page-title-main">Voltammetry</span> Method of analyzing electrochemical reactions

Voltammetry is a category of electroanalytical methods used in analytical chemistry and various industrial processes. In voltammetry, information about an analyte is obtained by measuring the current as the potential is varied. The analytical data for a voltammetric experiment comes in the form of a voltammogram, which plots the current produced by the analyte versus the potential of the working electrode.

The limit of detection is the lowest signal, or the lowest corresponding quantity to be determined from the signal, that can be observed with a sufficient degree of confidence or statistical significance. However, the exact threshold used to decide when a signal significantly emerges above the continuously fluctuating background noise remains arbitrary and is a matter of policy and often of debate among scientists, statisticians and regulators depending on the stakes in different fields.

In a chemical analysis, the internal standard method involves adding the same amount of a chemical substance to each sample and calibration solution. The internal standard responds proportionally to changes in the analyte and provides a similar, but not identical, measurement signal. It must also be absent from the sample matrix to ensure there is no other source of the internal standard present. Taking the ratio of analyte signal to internal standard signal and plotting it against the analyte concentrations in the calibration solutions will result in a calibration curve. The calibration curve can then be used to calculate the analyte concentration in an unknown sample.

Chemical imaging is the analytical capability to create a visual image of components distribution from simultaneous measurement of spectra and spatial, time information. Hyperspectral imaging measures contiguous spectral bands, as opposed to multispectral imaging which measures spaced spectral bands.

Multivariate optical computing, also known as molecular factor computing, is an approach to the development of compressed sensing spectroscopic instruments, particularly for industrial applications such as process analytical support. "Conventional" spectroscopic methods often employ multivariate and chemometric methods, such as multivariate calibration, pattern recognition, and classification, to extract analytical information from data collected at many different wavelengths. Multivariate optical computing uses an optical computer to analyze the data as it is collected. The goal of this approach is to produce instruments which are simple and rugged, yet retain the benefits of multivariate techniques for the accuracy and precision of the result.

<i>Journal of Chemometrics</i> Academic journal

The Journal of Chemometrics is a monthly peer-reviewed scientific journal published since 1987 by John Wiley & Sons. It publishes original scientific papers, reviews, and short communications on fundamental and applied aspects of chemometrics. The current editor-in-chief is Age K. Smilde.

Flow injection analysis (FIA) is an approach to chemical analysis. It is accomplished by injecting a plug of sample into a flowing carrier stream. The principle is similar to that of Segmented Flow Analysis (SFA) but no air is injected into the sample or reagent streams..

The Unscrambler X is a commercial software product for multivariate data analysis, used for calibration of multivariate data which is often in the application of analytical data such as near infrared spectroscopy and Raman spectroscopy, and development of predictive models for use in real-time spectroscopic analysis of materials. The software was originally developed in 1986 by Harald Martens and later by CAMO Software.

Acoustic resonance spectroscopy (ARS) is a method of spectroscopy in the acoustic region, primarily the sonic and ultrasonic regions. ARS is typically much more rapid than HPLC and NIR. It is non destructive and requires no sample preparation as the sampling waveguide can simply be pushed into a sample powder/liquid or in contact with a solid sample. To date, the AR spectrometer has successfully differentiated and quantified sample analytes in various forms;. It has been used to measure and monitor the progression of chemical reactions, such as the setting and hardening of concrete from cement paste to solid. Acoustic spectrometry has also been used to measure the volume fraction of colloids in a dispersion medium, as well as for the investigation of physical properties of colloidal dispersions, such as aggregation and particle size distribution. Typically, these experiments are carried out with sinusoidal excitation signals and the experimental observation of signal attenuation. From a comparison of theoretical attenuation to experimental observation, the particle size distribution and aggregation phenomena are inferred.

Paul J. Gemperline is an American analytical chemist and chemometrician. He is a Distinguished Professor of Chemistry at East Carolina University (ECU) located in Greenville, North Carolina and has been the recipient of several scientific awards, including the 2003 Eastern Analytical Symposium Award in Chemometrics. He is author of more than 60 publications in the field of chemometrics. Dr. Gemperline served as Dean of the Graduate School at ECU from 2008 to 2022. He retired from ECU June 30, 2022 and is now professor emeritus.

Multiway data analysis is a method of analyzing large data sets by representing a collection of observations as a multiway array, . The proper choice of data organization into (C+1)-way array, and analysis techniques can reveal patterns in the underlying data undetected by other methods.

Bruce R. Kowalski was an American professor of analytical chemistry who is acknowledged by the world-wide scientific community to be one of the founders of the field of chemometrics. He was the founding editor of Journal of Chemometrics, and the founding director of the Center for Process Analytical Chemistry at University of Washington in Seattle. Kowalski and Svante Wold formed the Chemometrics Society, which would later become the International Chemometrics Society.

<span class="mw-page-title-main">Tormod Næs</span> Norwegian statistician studying chemometrics

Tormod Næs is a Norwegian statistician working in chemometrics and sensometrics. He studies multivariate statistical analysis, spectroscopy, food science, and sensory science. His impact on chemometrics is exemplified by the over 8,000 citations to his most well-known book, Multivariate Calibration, and the awards in chemometrics that he has received.

References

  1. As recounted in Wold, S. (1995). "Chemometrics; what do we mean with it, and what do we want from it?". Chemometrics and Intelligent Laboratory Systems. 30 (1): 109–115. doi:10.1016/0169-7439(95)00042-9.
  2. Kowalski, Bruce R. (1975). "Chemometrics: Views and Propositions". J. Chem. Inf. Comput. Sci. 15 (4): 201–203. doi:10.1021/ci60004a002.
  3. Trygg, J.; Wold, S. (2003). "O2‐PLS, a two‐block (X–Y) latent variable regression (LVR) method with an integral OSC filter". Journal of Chemometrics. 17: 53–64. doi:10.1002/cem.775. S2CID   123071521.
  4. Malinowski, E. R.; Howery, D. G. (1980). Factor Analysis in Chemistry. New York: Wiley. ISBN   978-0471058816. (other editions followed in 1989, 1991 and 2002).
  5. Sharaf, M. A.; Illman, D. L.; Kowalski, B. R., eds. (1986). Chemometrics. New York: Wiley. ISBN   978-0471831068.
  6. Massart, D. L.; Vandeginste, B. G. M.; Deming, S. M.; Michotte, Y.; Kaufman, L. (1988). Chemometrics: a textbook. Amsterdam: Elsevier. ISBN   978-0444426604.
  7. 1 2 Martens, H.; Naes, T. (1989). Multivariate Calibration. New York: Wiley. ISBN   978-0471909798.
  8. Geladi, P.; Esbensen, K. (2005). "The Start and Early History of Chemometrics: Selected Interviews. Part 1". J. Chemometrics . 4 (5): 337–354. doi:10.1002/cem.1180040503. S2CID   120490459.
  9. Esbensen, K.; Geladi, P. (2005). "The Start and Early History of Chemometrics: Selected Interviews. Part 2". J. Chemometrics . 4 (6): 389–412. doi:10.1002/cem.1180040604. S2CID   221546473.
  10. Barton, Bastian; Thomson, James; Lozano Diz, Enrique; Portela, Raquel (September 2022). "Chemometrics for Raman Spectroscopy Harmonization". Applied Spectroscopy. 76 (9): 1021–1041. Bibcode:2022ApSpe..76.1021B. doi:10.1177/00037028221094070. ISSN   0003-7028. PMID   35622984. S2CID   249129065.
  11. Franke, J. (2002). "Inverse Least Squares and Classical Least Squares Methods for Quantitative Vibrational Spectroscopy". In Chalmers, John M (ed.). Handbook of Vibrational Spectroscopy. New York: Wiley. doi:10.1002/0470027320.s4603. ISBN   978-0471988472.
  12. Brown, C. D. (2004). "Discordance between Net Analyte Signal Theory and Practical Multivariate Calibration". Analytical Chemistry. 76 (15): 4364–4373. doi:10.1021/ac049953w. PMID   15283574.
  13. Krutchkoff, R. G. (1969). "Classical and inverse regression methods of calibration in extrapolation". Technometrics. 11 (3): 11–15. doi:10.1080/00401706.1969.10490714.
  14. Hunter, W. G. (1984). "Statistics and chemistry, and the linear calibration problem". In Kowalski, B. R. (ed.). Chemometrics: mathematics and statistics in chemistry. Boston: Riedel. ISBN   978-9027718464.
  15. Tellinghuisen, J. (2000). "Inverse vs. classical calibration for small data sets". Fresenius' J. Anal. Chem. 368 (6): 585–588. doi:10.1007/s002160000556. PMID   11228707. S2CID   21166415.
  16. Oliveri, Paolo (2017). "Class-modelling in food analytical chemistry: Development, sampling, optimisation and validation issues – A tutorial". Analytica Chimica Acta. 982: 9–19. doi:10.1016/j.aca.2017.05.013. hdl: 11567/881059 . PMID   28734370.
  17. Lawton, W. H.; Sylvestre, E. A. (1971). "Self Modeling Curve Resolution". Technometrics. 13 (3): 617–633. doi:10.1080/00401706.1971.10488823.
  18. Sylvestre, E. A.; Lawton, W. H.; Maggio, M. S. (1974). "Curve Resolution Using a Postulated Chemical Reaction". Technometrics. 16 (3): 353–368. doi:10.1080/00401706.1974.10489204.
  19. de Juan, A.; Tauler, R. (2003). "Chemometrics Applied to Unravel Multicomponent Processes and Mixtures. Revisiting Latest Trends in Multivariate Resolution". Analytica Chimica Acta. 500 (1–2): 195–210. doi:10.1016/S0003-2670(03)00724-4.
  20. de Juan, A.; Tauler, R. (2006). "Multivariate Curve Resolution (MCR) from 2000: Progress in Concepts and Applications". Critical Reviews in Analytical Chemistry. 36 (3–4): 163–176. doi:10.1080/10408340600970005. S2CID   95309963.
  21. Deming, S. N.; Morgan, S. L. (1987). Experimental design: a chemometric approach. Elsevier. ISBN   978-0444427342.
  22. Bruns, R. E.; Scarminio, I. S.; de Barros Neto, B. (2006). Statistical design – chemometrics. Amsterdam: Elsevier. ISBN   978-0444521811.
  23. Wentzell, P. D.; Brown, C. D. (2000). "Signal Processing in Analytical Chemistry". In Meyers, R. A. (ed.). Encyclopedia of Analytical Chemistry. Wiley. pp. 9764–9800.
  24. Oliveri, Paolo; Malegori, Cristina; Simonetti, Remo; Casale, Monica (2019). "The impact of signal pre-processing on the final interpretation of analytical outcomes – A tutorial". Analytica Chimica Acta. 1058: 9–17. doi:10.1016/j.aca.2018.10.055. PMID   30851858. S2CID   73727614.
  25. Olivieri, A. C.; Faber, N. M.; Ferre, J.; Boque, R.; Kalivas, J. H.; Mark, H. (2006). "Guidelines for calibration in analytical chemistry Part 3. Uncertainty estimation and figures of merit for multivariate calibration". Pure and Applied Chemistry. 78 (3): 633–650. doi: 10.1351/pac200678030633 . S2CID   50546210.
  26. Illman, D. L.; Callis, J. B.; Kowalski, B. R. (1986). "Process Analytical Chemistry: a new paradigm for analytical chemists". American Laboratory. 18: 8–10.
  27. MacGregor, J. F.; Kourti, T. (1995). "Statistical control of multivariate processes". Control Engineering Practice. 3 (3): 403–414. doi:10.1016/0967-0661(95)00014-L.
  28. Martin, E. B.; Morris, A. J. (1996). "An overview of multivariate statistical process control in continuous and batch process performance monitoring". Transactions of the Institute of Measurement & Control. 18 (1): 51–60. doi:10.1177/014233129601800107. S2CID   120516715.
  29. Hirschfeld, T.; Callis, J. B.; Kowalski, B. R. (1984). "Chemical sensing in process analysis". Science . 226 (4672): 312–318. Bibcode:1984Sci...226..312H. doi:10.1126/science.226.4672.312. PMID   17749872. S2CID   38093353.
  30. Smilde, A. K.; Bro, R.; Geladi, P. (2004). Multi-way analysis with applications in the chemical sciences. Wiley.
  31. Bro, R.; Workman, J. J.; Mobley, P. R.; Kowalski, B. R. (1997). "Overview of chemometrics applied to spectroscopy: 1985–95, Part 3—Multiway analysis". Applied Spectroscopy Reviews. 32 (3): 237–261. Bibcode:1997ApSRv..32..237B. doi:10.1080/05704929708003315.

Further reading