Mathematical statistics with Mathematica
The present book proposes a professional approach to the statistical theory from the computational and mathematical point of view, using the powerful computer software Mathematica as support for the advanced numeric and symbolic computations. Mathematica is not just a specialized statistical package (like SPSS, Gauss, Systat, SAS, S-Plus, etc.), but much more: Mathematica provides both a (arbitrary-precision) numerical engine and a powerful symbolic/algebraic engine oriented to problem solving in mathematical statistics. Mathematica is running on a wide variety of computer platforms; the present book is bundled with the special software package mathStatica, an add-on to Mathematica, a sophisticated toolset specially designed for doing mathematical statistics (enclosed on the two attached CD-ROMs). Addressed at researchers, students, and professional practitioners with a good knowledge of mathematical statistics, the focus of this book is on problem solving rather than mathematical theorems and proofs, using the mathStatica and Mathematica packages to bring mathematical statistics to real life. The contents of the book are structured within 12 chapters, covering from elementary up to the highest-level topics, results and problems. Chapter 1 (Introduction) tells the reader how to install the enclosed Mathematica software, and provides the core functions, notations and conventions. Chapter 2 (Continuous Random Variables) shows mathStatica computing for various continuous measures of location (mean, mode, median and quantiles), moments and generating functions, conditional distributions and expectations, and pseudo-random number generation. Chapter 3 (Discrete Random Variables) turns the reader's attention toward experiments defined on countable (discrete) sample spaces. Various discrete random variables are examined, whose domain of support is the set of non-negative integers: Bernoulli, binomial, Poisson, geometric, negative binomial, and hyper-geometric distributions, as well as component-mixing and parameter-mixing distributions. Chapter 4 (Distributions of Functions of Random Variables) deals with an important practical problem: which is the distribution of a function applied to a random variable, whose density function is given? Two main approaches are experimented: the Transformation Method, and the MGF (moment generating function) Method. Chapter 5 (Systems of Distributions) discusses three systems of distributions: (1) the Pearson family, which defines a density in terms of its slope; (2) the Johnson system, describing density in terms of transformations of the standard normal; (3) Gram-Charlier expansions which represent a density as a series expansion of the standard normal density. Chapter 6 (Mulivariate Distributions) provides the transition from univariate to multivariate statistical analysis, introducing and experimenting new concepts such as: joint densities, joint distribution functions, marginal and conditional distributions, multivariate expectations, product moments, etc., properties of dependence and independence, important multivariate distributions. Chapter 7 (Moments of Sampling Distributions) addresses two topics: (a) unbiased estimators of population moments, and (b) population moments of sample moments. Chapter 8 (Asymptotic Theory) investigates desirable asymptotic properties such as consistency and limiting normality, in the context of sample mean and sample sum (in the simple case of i.i.d. random variables). Chapter 9 (Statistical Decision Theory) examines how to select an estimator, in an optimal fashion, from the set of decisions. Loss and risk, mean square error as risk, and order statistics are the topics of computational experiments. In Chapter 10 (Unbiased Parameter Estimation), the statistical information that is present in a parameter in a statistical model is measured using Fisher information, sample information, etc., this leading to Cramér-Rao lower lower bound and best unbiased estimators. The theoretical development of Minimum Variance Unbiased Estimation (MVUE) is based on the famous theorem of Rao and Blackwell, in the context of MVUE sufficient statistics and its completeness. Chapter 11 (Principles of Maximum Likelihood Estimation) and Chapter 12 (Maximum Likelihood Estimation in Practice) propose interesting theoretical and applicational experiments related to ML (Maximum Likelihood) estimation based on the algorithms enclosed in the Mathematica and mathStatica packages. A useful Appendix ends the book. The bunch of exercises at the end of each chapter represents another argument to consider this book also a perfect companion to any kind of (universitary) course level in mathematical statistics.
- A unifying framework for \(k\)-statistics, polykays and their multivariate generalizations
- Using mathematical expressions in a statistical language
- Symbolic computation of moments of sampling distributions
- A partial-postponement decision cost model
- Distribution-dependent and distribution-free confidence intervals for the variance
- Using Fisher scoring to fit extended Poisson process models
- mathStatica
- On the success probability of solving unique SVP via BKZ
- Dual-tree complex wavelet coefficient magnitude modeling using scale mixtures of Rayleigh distribution for image denoising
- Quantifying uncertain system outputs via the multilevel Monte Carlo method. I: Central moment estimation
- Finite form representations for Meijer \(G\) and Fox \(H\) functions. Applied to multivariate likelihood ratio tests using Mathematica, MAXIMA and R
- Impact of bacterial hyperinfectivity on cholera epidemics in a spatially heterogeneous environment
- Accurate estimation with one order statistic
- A simulation study on classic and robust variable selection in linear regression
- Nash equilibria of Cauchy-random zero-sum and coordination matrix games
- Automating bivariate transformations
- Asymptotic expansions of the null distributions of discrepancy functions for general covariance structures under nonnormality
- \textbf{mathStatica}: mathematical statistics with Mathematica
- Cross ranking of cities and regions: population versus income
- Computer Algebra Derivation of the Bias of Linear Estimators of Autoregressive Models
- Mathematical methods and symbolic calculation in chemistry and chemical biology—The gathering momentum
- Higher Order Asymptotic Cumulants of Studentized Estimators in Covariance Structures
- scientific article; zbMATH DE number 1266757 (Why is no real title available?)
- scientific article; zbMATH DE number 1958335 (Why is no real title available?)
- Optimal unbiased estimation of some population central moments
- Computer algebra and algorithms for unbiased moment estimation of arbitrary order
- Difference-based variance estimation in nonparametric regression with repeated measurement data
- On higher-order moment and cumulant estimation
- Chaotic coexistence in a resource-consumer model
- A Viable Alternative to Resorting to Statistical Tables
- The accuracy of \textit{Mathematica 4} as a statistical package
- Nonlinear finite-time Lyapunov exponent and predictability
- On estimating the finite population variance by utilizing auxiliary information
- Canonical decomposition of belief functions based on Teugels' representation of the multivariate Bernoulli distribution
- Vector generalized linear and additive extreme value models
- Distribution modeling and simulation of gene expression data
This page was built for publication: Mathematical statistics with \(Mathematica\)
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2783518)
