Book contents
- Frontmatter
- Dedication
- Contents
- Preface to the Second Edition
- Preface to the First Edition
- Acknowledgments for the Second Edition
- Acknowledgments for the First Edition
- 1 Some Essential Notation
- 2 Signals, Integrals, and Sets of Measure Zero
- 3 The Inner Product
- 4 The Space L2 of Energy-Limited Signals
- 5 Convolutions and Filters
- 6 The Frequency Response of Filters and Bandlimited Signals
- 7 Passband Signals and Their Representation
- 8 Complete Orthonormal Systems and the Sampling Theorem
- 9 Sampling Real Passband Signals
- 10 Mapping Bits to Waveforms
- 11 Nyquist's Criterion
- 12 Stochastic Processes: Definition
- 13 Stationary Discrete-Time Stochastic Processes
- 14 Energy and Power in PAM
- 15 Operational Power Spectral Density
- 16 Quadrature Amplitude Modulation
- 17 Complex Random Variables and Processes
- 18 Energy, Power, and PSD in QAM
- 19 The Univariate Gaussian Distribution
- 20 Binary Hypothesis Testing
- 21 Multi-Hypothesis Testing
- 22 Sufficient Statistics
- 23 The Multivariate Gaussian Distribution
- 24 Complex Gaussians and Circular Symmetry
- 25 Continuous-Time Stochastic Processes
- 26 Detection in White Gaussian Noise
- 27 Noncoherent Detection and Nuisance Parameters
- 28 Detecting PAM and QAM Signals in White Gaussian Noise
- 29 Linear Binary Block Codes with Antipodal Signaling
- 30 The Radar Problem
- 31 A Glimpse at Discrete-Time Signal Processing
- 32 Intersymbol Interference
- A On the Fourier Series
- B On the Discrete-Time Fourier Transform
- C Positive Definite Functions
- D The Baseband Representation of Passband Stochastic Processes
- Bibliography
- Theorems Referenced by Name
- Abbreviations
- List of Symbols
- Index
22 - Sufficient Statistics
Published online by Cambridge University Press: 02 March 2017
- Frontmatter
- Dedication
- Contents
- Preface to the Second Edition
- Preface to the First Edition
- Acknowledgments for the Second Edition
- Acknowledgments for the First Edition
- 1 Some Essential Notation
- 2 Signals, Integrals, and Sets of Measure Zero
- 3 The Inner Product
- 4 The Space L2 of Energy-Limited Signals
- 5 Convolutions and Filters
- 6 The Frequency Response of Filters and Bandlimited Signals
- 7 Passband Signals and Their Representation
- 8 Complete Orthonormal Systems and the Sampling Theorem
- 9 Sampling Real Passband Signals
- 10 Mapping Bits to Waveforms
- 11 Nyquist's Criterion
- 12 Stochastic Processes: Definition
- 13 Stationary Discrete-Time Stochastic Processes
- 14 Energy and Power in PAM
- 15 Operational Power Spectral Density
- 16 Quadrature Amplitude Modulation
- 17 Complex Random Variables and Processes
- 18 Energy, Power, and PSD in QAM
- 19 The Univariate Gaussian Distribution
- 20 Binary Hypothesis Testing
- 21 Multi-Hypothesis Testing
- 22 Sufficient Statistics
- 23 The Multivariate Gaussian Distribution
- 24 Complex Gaussians and Circular Symmetry
- 25 Continuous-Time Stochastic Processes
- 26 Detection in White Gaussian Noise
- 27 Noncoherent Detection and Nuisance Parameters
- 28 Detecting PAM and QAM Signals in White Gaussian Noise
- 29 Linear Binary Block Codes with Antipodal Signaling
- 30 The Radar Problem
- 31 A Glimpse at Discrete-Time Signal Processing
- 32 Intersymbol Interference
- A On the Fourier Series
- B On the Discrete-Time Fourier Transform
- C Positive Definite Functions
- D The Baseband Representation of Passband Stochastic Processes
- Bibliography
- Theorems Referenced by Name
- Abbreviations
- List of Symbols
- Index
Summary
Introduction
In layman's terms, a sufficient statistic for guessing M based on the observable Y is a random variable or a collection of random variables that contains all the information in Y that is relevant for guessing M. This is a particularly useful concept when the sufficient statistic is more concise than the observables. For example, if we observe the results of a thousand coin tosses Y1, …, Y1000 and we wish to test whether the coin is fair or has a bias of 1/4, then a sufficient statistic turns out to be the number of “heads”among the outcomes Y1, …, Y1000. Another example was encountered in Section 20.12. There the observable was a two-dimensional random vector, and the sufficient statistic summarized the information that was relevant for guessing H in a scalar random variable; see (20.69).
In this chapter we provide a formal definition of sufficient statistics in the multihypothesis setting and explore the concept in some detail. We shall see that our definition is compatible with Definition 20.12.2, which we gave for the binary case. We only address the case where the observations take value in the d-dimensional Euclidean space Rd. Also, we only treat the case of guessing among a finite number of alternatives. We thus consider a finite set of messages
whereM ≥ 2, and we assume that associated with each message is a density on Rd, i.e., a nonnegative Borel measurable function that integrates to one.
The concept of sufficient statistics is defined for the family of densities
it is unrelated to a prior. But when we wish to use it in the context of hypothesis testing we need to introduce a probabilistic setting. If, in addition to the family, we introduce a prior, then we can discuss the pair (M,Y), where Pr[M = m] = πm, and where, conditionally on M = m, the distribution of Y is of density.
- Type
- Chapter
- Information
- A Foundation in Digital Communication , pp. 468 - 493Publisher: Cambridge University PressPrint publication year: 2017