Fulfilling the need for a practical user’s guide, Statistics in MATLAB: A Primer provides an accessible introduction to the latest version of MATLAB® and its extensive functionality for statistics. Assuming a basic knowledge of statistics and probability as well as a fundamental understanding of linear algebra concepts, this book: Covers capabilities in the main MATLAB package, the Statistics Toolbox, and the student version of MATLAB Presents examples of how MATLAB can be used to analyze data Offers access to a companion website with data sets and additional examples Contains figures and visual aids to assist in application of the software Explains how to determine what method should be used for analysis Statistics in MATLAB: A Primer is an ideal reference for undergraduate and graduate students in engineering, mathematics, statistics, economics, biostatistics, and computer science. It is also appropriate for a diverse professional market, making it a valuable addition to the libraries of researchers in statistics, computer science, data mining, machine learning, image analysis, signal processing, and engineering.
Based on a course taught by the author, this book combines the theoretical underpinnings of statistics with the practical analysis of Earth sciences data using MATLAB. The book is organized to introduce the underlying concepts, and then extends these to the data, covering methods that are most applicable to Earth sciences. Topics include classical parametric estimation and hypothesis testing, and more advanced least squares-based, nonparametric, and resampling estimators. Multivariate data analysis, not often encountered in introductory texts, is presented later in the book, and compositional data is treated at the end. Datasets and bespoke MATLAB scripts used in the book are available online, as well as additional datasets and suggested questions for use by instructors. Aimed at entering graduate students and practicing researchers in the Earth and ocean sciences, this book is ideal for those who want to learn how to analyse data using MATLAB in a statistically-rigorous manner.
A Strong Practical Focus on Applications and Algorithms Computational Statistics Handbook with MATLAB®, Third Edition covers today’s most commonly used techniques in computational statistics while maintaining the same philosophy and writing style of the bestselling previous editions. The text keeps theoretical concepts to a minimum, emphasizing the implementation of the methods. New to the Third Edition This third edition is updated with the latest version of MATLAB and the corresponding version of the Statistics and Machine Learning Toolbox. It also incorporates new sections on the nearest neighbor classifier, support vector machines, model checking and regularization, partial least squares regression, and multivariate adaptive regression splines. Web Resource The authors include algorithmic descriptions of the procedures as well as examples that illustrate the use of algorithms in data analysis. The MATLAB code, examples, and data sets are available online.
A coherent, concise and comprehensive course in the statistics needed for a modern career in chemical engineering; covers all of the concepts required for the American Fundamentals of Engineering examination. This book shows the reader how to develop and test models, design experiments and analyse data in ways easily applicable through readily available software tools like MS Excel® and MATLAB®. Generalized methods that can be applied irrespective of the tool at hand are a key feature of the text. The reader is given a detailed framework for statistical procedures covering: · data visualization; · probability; · linear and nonlinear regression; · experimental design (including factorial and fractional factorial designs); and · dynamic process identification. Main concepts are illustrated with chemical- and process-engineering-relevant examples that can also serve as the bases for checking any subsequent real implementations. Questions are provided (with solutions available for instructors) to confirm the correct use of numerical techniques, and templates for use in MS Excel and MATLAB can also be downloaded from extras.springer.com. With its integrative approach to system identification, regression and statistical theory, Statistics for Chemical and Process Engineers provides an excellent means of revision and self-study for chemical and process engineers working in experimental analysis and design in petrochemicals, ceramics, oil and gas, automotive and similar industries and invaluable instruction to advanced undergraduate and graduate students looking to begin a career in the process industries.
Exploring roles critical to environmental toxicology, Modeling and Simulation in Ecotoxicology with Applications in MATLAB® and Simulink® covers the steps in modeling and simulation from problem conception to validation and simulation analysis. Using the MATLAB and Simulink programming languages, the book presents examples of mathematical functions and simulations, with special emphasis on how to develop mathematical models and run computer simulations of ecotoxicological processes. Designed for students and professionals with little or no experience in modeling, the book includes: General principles of modeling and simulation and an introduction to MATLAB and Simulink Stochastic modeling where variability and uncertainty are acknowledged by making parameters random variables Toxicological processes from the level of the individual organism, with worked examples of process models in either MATLAB or Simulink Toxicological processes at the level of populations, communities, and ecosystems Parameter estimation using least squares regression methods The design of simulation experiments similar to the experimental design applied to laboratory or field experiments Methods of postsimulation analysis, including stability analysis and sensitivity analysis Different levels of model validation and how they are related to the modeling purpose The book also provides three individual case studies. The first involves a model developed to assess the relative risk of mortality following exposure to insecticides in different avian species. The second explores the role of diving behavior on the inhalation and distribution of oil spill naphthalene in bottlenose dolphins. The final case study looks at the dynamics of mercury in Daphnia that are exposed to simulated thermal plumes from a hypothetical power plant cooling system. Presented in a rigorous yet accessible style, the methodology is versatile enough to be readily applicable not only to environmental toxicology but a range of other biological fields.
As with the bestselling first edition, Computational Statistics Handbook with MATLAB®, Second Edition covers some of the most commonly used contemporary techniques in computational statistics. With a strong, practical focus on implementing the methods, the authors include algorithmic descriptions of the procedures as well as examples that illustrate the use of the algorithms in data analysis. Updated for MATLAB® R2007a and the Statistics Toolbox, Version 6.0, this edition incorporates many additional computational statistics topics. New to the Second Edition • New functions for multivariate normal and multivariate t distributions • Updated information on the new MATLAB functionality for univariate and bivariate histograms, glyphs, and parallel coordinate plots • New content on independent component analysis, nonlinear dimensionality reduction, and multidimensional scaling • New topics on linear classifiers, quadratic classifiers, and voting methods, such as bagging, boosting, and random forests • More methods for unsupervised learning, including model-based clustering and techniques for assessing the results of clustering • A new chapter on parametric models that covers spline regression models, logistic regression, and generalized linear models • Expanded information on smoothers, such as bin smoothing, running mean and line smoothers, and smoothing splines With numerous problems and suggestions for further reading, this accessible text facilitates an understanding of computational statistics concepts and how they are employed in data analysis.
Exploratory data analysis (EDA) was conceived at a time when computers were not widely used, and thus computational ability was rather limited. As computational sophistication has increased, EDA has become an even more powerful process for visualizing and summarizing data before making model assumptions to generate hypotheses, encompassing larger and more complex data sets. There are many resources for those interested in the theory of EDA, but this is the first book to use MATLAB to illustrate the computational aspects of this discipline. Exploratory Data Analysis with MATLAB presents the methods of EDA from a computational perspective. The authors extensively use MATLAB code and algorithm descriptions to provide state-of-the-art techniques for finding patterns and structure in data. Addressing theory, they also incorporate many annotated references to direct readers to the more theoretical aspects of the methods. The book presents an approach using the basic functions from MATLAB and the MATLAB Statistics Toolbox, in order to be more accessible and enduring. It also contains pseudo-code to enable users of other software packages to implement the algorithms. This text places the tools needed to implement EDA theory at the fingertips of researchers, applied mathematicians, computer scientists, engineers, and statisticians by using a practical/computational approach.
Through its scope and depth of coverage, this book addresses the needs of the vibrant and rapidly growing engineering fields, bioengineering and biomedical engineering, while implementing software that engineers are familiar with. The author integrates introductory statistics for engineers and introductory biostatistics as a single textbook heavily oriented to computation and hands on approaches. For example, topics ranging from the aspects of disease and device testing, Sensitivity, Specificity and ROC curves, Epidemiological Risk Theory, Survival Analysis, or Logistic and Poisson Regressions are covered. In addition to the synergy of engineering and biostatistical approaches, the novelty of this book is in the substantial coverage of Bayesian approaches to statistical inference. Many examples in this text are solved using both the traditional and Bayesian methods, and the results are compared and commented.
Probability distributions are theoretical distributions based on assumptions about asource population. The distributions assign probability to the event that a randomvariable has a specific, discrete value, or falls within a specified range of continuous values. Statistics and Machine Learning Toolbox offers several ways to work with probabilitydistributions. -Use "Probability Distribution Objects" to fit a probability distributionobject to sample data, or to create a probability distribution object with specifiedparameter values. -Use "Probability Distribution Functions" to work with data input frommatrices, tables, and dataset arrays. -Use "Probability Distribution Apps and User Interfaces" to interactively fit, explore, and generate random numbers from probabilitydistributions. Available apps and user interfaces include: -The Distribution Fitter app -The Probability Distribution Function user interface -The Random Number Generation user interface (randtool) Pseudorandom numbers are generated by deterministic algorithms. They are "random" in the sense that, on average, they pass statistical tests regarding their distribution andcorrelation. They differ from true random numbers in that they are generated by analgorithm, rather than a truly random process.Random number generators (RNGs) like those in MATLAB are algorithms for generatingpseudorandom numbers with a specified distribution. Methods for generating pseudorandom numbers usually start with uniform randomnumbers, like the MATLAB rand function produces, but MATLAB produce random numbers from other distributions. This book develops the work with probability distribution and random numbers across examples using MATLAB. The more important topics in the book are de next: -Working with Probability Distributions -Continuous Distributions (Data) -Continuous Distributions (Statistics) -Discrete Distributions -Multivariate Distributions -Nonparametric Distributions -Flexible Distribution Families -Maximum Likelihood Estimation -Negative Loglikelihood Functions -Random Number Generation -Nonparametric and Empirical Probability Distributions -Fit Kernel Distribution Object to Data -Fit Kernel Distribution Using ksdensity -Fit Distributions to Grouped Data Using ksdensity -Fit a Nonparametric Distribution with Pareto Tails -Generate Random Numbers Using the TriangularDistribution -Model Data Using the Distribution Fitter App -Fit a Distribution Using the Distribution Fitter App -Custom Distributions Using the Distribution Fitter App -Explore the Random Number Generation UI -Compare Multiple Distribution Fits -Fit Probability Distribution Objects to Grouped Data -Multinomial Probability Distribution Objects -Multinomial Probability Distribution Functions -Generate Random Numbers Using Uniform Distribution Inversion -Represent Cauchy Distribution Using t Location-Scale -Generate Cauchy Random Numbers Using Student's t -Generate Correlated Data Using Rank Correlation -Gaussian Mixture Models -Create a Gaussian Mixture Model -Fit a Gaussian Mixture Model to Data -Simulate Data from a Gaussian Mixture Model -Copulas: Generate Correlated Samples -Generating Random Data -Random Number Generation Functions -Common Pseudorandom Number Generation Methods -Generating Quasi-Random Numbers -Generating Data Using Flexible Families of Distributions -Bayesian Linear Regression Using Hamiltonian MonteCarlo