Limit search to available items
Book Cover
E-book
Author Monahan, John F.

Title Numerical methods of statistics / John F. Monahan
Edition 2nd ed
Published Cambridge ; New York : Cambridge University Press, 2011

Copies

Description 1 online resource (xvi, 447 pages) : illustrations
Series Cambridge series in statistical and probabilistic mathematics ; [32]
Cambridge series on statistical and probabilistic mathematics ; 32.
Contents 1. Algorithms and Computers -- 1.1. Introduction -- 1.2. Computers -- 1.3. Software and Computer Languages -- 1.4. Data Structures -- 1.5. Programming Practice -- 1.6. Some Comments on R -- References -- 2. Computer Arithmetic -- 2.1. Introduction -- 2.2. Positional Number Systems -- 2.3. Fixed Point Arithmetic -- 2.4. Floating Point Representations -- 2.5. Living with Floating Point Inaccuracies -- 2.6. The Pale and Beyond -- 2.7. Conditioned Problems and Stable Algorithms -- Programs and Demonstrations -- Exercises -- References -- 3. Matrices and Linear Equations -- 3.1. Introduction -- 3.2. Matrix Operations -- 3.3. Solving Triangular Systems -- 3.4. Gaussian Elimination -- 3.5. Cholesky Decomposition -- 3.6. Matrix Norms -- 3.7. Accuracy and Conditioning -- 3.8. Matrix Computations in R -- Programs and Demonstrations -- Exercises -- References
4. More Methods for Solving Linear Equations -- 4.1. Introduction -- 4.2. Full Elimination with Complete Pivoting -- 4.3. Banded Matrices -- 4.4. Applications to ARMA Time-Series Models -- 4.5. Toeplitz Systems -- 4.6. Sparse Matrices -- 4.7. Iterative Methods -- 4.8. Linear Programming -- Programs and Demonstrations -- Exercises -- References -- 5. Regression Computations -- 5.1. Introduction -- 5.2. Condition of the Regression Problem -- 5.3. Solving the Normal Equations -- 5.4. Gram-Schmidt Orthogonalization -- 5.5. Householder Transformations -- 5.6. Householder Transformations for Least Squares -- 5.7. Givens Transformations -- 5.8. Givens Transformations for Least Squares -- 5.9. Regression Diagnostics -- 5.10. Hypothesis Tests -- 5.11. Conjugate Gradient Methods -- 5.12. Doolittle, the Sweep, and All Possible Regressions -- 5.13. Alternatives to Least Squares -- 5.14. Comments -- Programs and Demonstrations -- Exercises -- References
6. Eigenproblems -- 6.1. Introduction -- 6.2. Theory -- 6.3. Power Methods -- 6.4. The Symmetric Eigenproblem and Tridiagonalization -- 6.5. The QR Algorithm -- 6.6. Singular Value Decomposition -- 6.7. Applications -- 6.8. Complex Singular Value Decomposition -- Programs and Demonstrations -- Exercises -- References -- 7. Functions: Interpolation, Smoothing, and Approximation -- 7.1. Introduction -- 7.2. Interpolation -- 7.3. Interpolating Splines -- 7.4. Curve Fitting with Splines: Smoothing and Regression -- 7.5. Mathematical Approximation -- 7.6. Practical Approximation Techniques -- 7.7. Computing Probability Functions -- Programs and Demonstrations -- Exercises -- References -- 8. Introduction to Optimization and Nonlinear Equations -- 8.1. Introduction -- 8.2. Safe Univariate Methods: Lattice Search, Golden Section, and Bisection -- 8.3. Root Finding -- 8.4. First Digression: Stopping and Condition
8.5. Multivariate Newton's Methods -- 8.6. Second Digression: Numerical Differentiation -- 8.7. Minimization and Nonlinear Equations -- 8.8. Condition and Scaling -- 8.9. Implementation -- 8.10. A Non-Newton Method: Nelder-Mead -- Programs and Demonstrations -- Exercises -- References -- 9. Maximum Likelihood and Nonlinear Regression -- 9.1. Introduction -- 9.2. Notation and Asymptotic Theory of Maximum Likelihood -- 9.3. Information, Scoring, and Variance Estimates -- 9.4. An Extended Example -- 9.5. Concentration, Iteration, and the EM Algorithm -- 9.6. Multiple Regression in the Context of Maximum Likelihood -- 9.7. Generalized Linear Models -- 9.8. Nonlinear Regression -- 9.9. Parameterizations and Constraints -- Programs and Demonstrations -- Exercises -- References -- 10. Numerical Integration and Monte Carlo Methods -- 10.1. Introduction -- 10.2. Motivating Problems -- 10.3. One-Dimensional Quadrature
10.4. Numerical Integration in Two or More Variables -- 10.5. Uniform Pseudorandom Variables -- 10.6. Quasi-Monte Carlo Integration -- 10.7. Strategy and Tactics -- Programs and Demonstrations -- Exercises -- References -- 11. Generating Random Variables from Other Distributions -- 11.1. Introduction -- 11.2. General Methods for Continuous Distributions -- 11.3. Algorithms for Continuous Distributions -- 11.4. General Methods for Discrete Distributions -- 11.5. Algorithms for Discrete Distributions -- 11.6. Other Randomizations -- 11.7. Accuracy in Random Number Generation -- Programs and Demonstrations -- Exercises -- References -- 12. Statistical Methods for Integration and Monte Carlo -- 12.1. Introduction -- 12.2. Distribution and Density Estimation -- 12.3. Distributional Tests -- 12.4. Importance Sampling and Weighted Observations -- 12.5. Testing Importance Sampling Weights -- 12.6. Laplace Approximations
12.7. Randomized Quadrature -- 12.8. Spherical-Radial Methods -- Programs and Demonstrations -- Exercises -- References -- 13. Markov Chain Monte Carlo Methods -- 13.1. Introduction -- 13.2. Markov Chains -- 13.3. Gibbs Sampling -- 13.4. Metropolis-Hastings Algorithm -- 13.5. Time-Series Analysis -- 13.6. Adaptive Acceptance/Rejection -- 13.7. Diagnostics -- Programs and Demonstrations -- Exercises -- References -- 14. Sorting and Fast Algorithms -- 14.1. Introduction -- 14.2. Divide and Conquer -- 14.3. Sorting Algorithms -- 14.4. Fast Order Statistics and Related Problems -- 14.5. Fast Fourier Transform -- 14.6. Convolutions and the Chirp-z Transform -- 14.7. Statistical Applications of the FFT -- 14.8. Combinatorial Problems -- Programs and Demonstrations -- Exercises -- References
Summary This book explains how computer software is designed to perform the tasks required for sophisticated statistical analysis. For statisticians, it examines the nitty-gritty computational problems behind statistical methods. For mathematicians and computer scientists, it looks at the application of mathematical tools to statistical problems. The first half of the book offers a basic background in numerical analysis that emphasizes issues important to statisticians. The next several chapters cover a broad array of statistical tools, such as maximum likelihood and nonlinear regression. The author also treats the application of numerical tools; numerical integration and random number generation are explained in a unified manner reflecting complementary views of Monte Carlo methods. Each chapter contains exercises that range from simple questions to research problems. Most of the examples are accompanied by demonstration and source code available from the author's website. New in this second edition are demonstrations coded in R, as well as new sections on linear programming and the Nelder-Mead search algorithm
Bibliography Includes bibliographical references and indexes
Notes English
Print version record
Subject Mathematical statistics -- Data processing.
Numerical analysis.
MATHEMATICS -- Probability & Statistics -- General.
Mathematical statistics -- Data processing
Numerical analysis
Form Electronic book
LC no. 2011287063
ISBN 9781139082112
1139082116
9781139079846
1139079840
9781139077552
1139077554
9780511977176
0511977174
1107213894
9781107213890
1283112329
9781283112321
9786613112323
6613112321
1139075292
9781139075299
1139069527
9781139069526