rent-now

Rent More, Save More! Use code: ECRENTAL

5% off 1 book, 7% off 2 books, 10% off 3+ books

9780387987613

Conditional Specification of Statistical Models

by ; ;
  • ISBN13:

    9780387987613

  • ISBN10:

    0387987614

  • Format: Hardcover
  • Copyright: 1999-11-01
  • Publisher: Springer Verlag
  • Purchase Benefits
  • Free Shipping Icon Free Shipping On Orders Over $35!
    Your order must be $35 or more to qualify for free economy shipping. Bulk sales, PO's, Marketplace items, eBooks and apparel do not qualify for this offer.
  • eCampus.com Logo Get Rewarded for Ordering Your Textbooks! Enroll Now
List Price: $169.99 Save up to $134.35
  • Digital
    $77.22
    Add to Cart

    DURATION
    PRICE

Summary

The concept of conditional specification of distributions is not new but, except in normal families, it has not been well developed in the literature. Computational difficulties undoubtedly hindered or discouraged developments in this direction. However, such roadblocks are of dimished importance today. Questions of compatibility of conditional and marginal specifications of distributions are of fundamental importance in modeling scenarios. Models with conditionals in exponential families are particularly tractable and provide useful models in a broad variety of settings.

Table of Contents

Preface vii
Conditional Specification: Concepts and Theorems
1(18)
Why Conditional Specified Models?
1(1)
How May One Specify a Bivariate Distribution?
2(2)
Early Work on Conditionally Specified Models
4(1)
The Conditional Specification Paradigm
5(1)
Compatible Conditionals: Finite Discrete Case
5(3)
Compatibility in More General Settings
8(3)
Uniqueness
11(1)
Conditionals in Prescribed Families
12(2)
An Example
14(2)
Bibliographic Notes
16(3)
Exercises
16(3)
Exact and Near Compatibility
19(34)
Introduction
19(1)
Review and Extensions of Compatibility Results
19(11)
Near Compatibility
30(1)
Minimal Incompatibility in Terms of Kullback-Leibler Pseudo-Distance
31(5)
More Than One Expert
36(1)
Related Discrepancy Measures
37(4)
Markovian Measures of Discrepancy
41(2)
ε-Compatibility
43(7)
Extensions of More General Settings
50(1)
Bibliographic Notes
50(3)
Exercises
51(2)
Distributions with Normal Conditionals
53(22)
Introduction
53(1)
Variations on the Classical Bivariate Normal Theme
53(2)
Normal Conditionals
55(4)
Properties of the Normal Conditionals Distribution
59(6)
The Centered Model
65(7)
Bibliographic Notes
72(3)
Exercises
72(3)
Conditionals in Exponential Families
75(28)
Introduction
75(1)
Distributions with Conditionals in Given Exponential Families
75(3)
Dependence in CEF Distributions
78(2)
Exponential Conditionals
80(2)
Normal Conditionals
82(1)
Gamma Conditionals
83(5)
Weibull Conditionals
88(1)
Gamma--Normal Conditionals
89(1)
Power-Function and Other Weighted Distributions as Conditionals
90(3)
Beta Conditionals
93(1)
Inverse Gaussian Conditionals
94(1)
Three Discrete Examples (Binomial, Geometric, and Poisson)
95(2)
Poisson--Gamma Conditionals
97(2)
Bibliographic Notes
99(4)
Exercises
99(4)
Other Conditionally Specified Families
103(30)
Introduction
103(1)
Bivariate Distributions with Pareto Conditionals
104(3)
Pearson Type VI Conditionals
107(2)
Bivariate Distributions with Generalized Pareto Conditionals
109(7)
Bivariate Distributions with Cauchy Conditionals
116(6)
Bivariate Distributions with Student-t Conditionals
122(1)
Bivariate Distributions with Uniform Conditionals
123(1)
Possibly Translated Exponential Conditionals
124(1)
Bivariate Distributions with Scaled Beta Conditionals
125(1)
Weibull and Logistic Conditionals
126(2)
Mixtures
128(1)
Bibliographic Notes
129(4)
Exercises
129(4)
Improper and Nonstandard Models
133(14)
Introduction
133(1)
Logistic Regression
134(1)
Uniform Conditionals
135(1)
Exponential and Weibull Conditionals
135(1)
Measurement Error Models
136(1)
Stochastic Processes and Wohler Fields
136(8)
Physical Interpretations of the Model Parameters
144(1)
Bibliographic Notes
145(2)
Exercises
145(2)
Characterizations Involving Conditional Moments
147(28)
Introduction
147(1)
Mardia's Bivariate Pareto Distribution
148(1)
Linear Regressions with Conditionals in Exponential Families
149(1)
Linear Regressions with Conditionals in Location Families
150(2)
Specified Regressions with Conditionals in Scale Families
152(1)
Conditionals in Location-Scale Families with Specified Moments
153(2)
Given One Conditional Distribution and the Other Regression
155(14)
Conditional Moments Only Given
169(1)
Bibliographic Notes
170(5)
Exercises
170(5)
Multivariate Extensions
175(22)
Introduction
175(1)
Extension by Underlining
175(1)
Compatibility in Three Dimensions
176(1)
Compatibility in Higher Dimensions
177(1)
Conditionals in Prescribed Families
177(1)
Conditionals in Exponential Families
178(1)
Multivariate Exponential Conditionals Distribution
179(1)
Multivariate Normal Conditionals Distribution
180(1)
Multivariate Cauchy Conditionals Distribution
181(1)
Multivariate Uniform Conditionals Distribution
181(1)
Multivariate Pareto Conditionals Distribution
182(1)
Multivariate Beta Conditionals Distribution
183(2)
Multivariate Binomial Conditionals Distribution
185(1)
Further Extension by Underlining
185(1)
Characterization of Multivariate Normality
186(4)
Multivariate Normality in More Abstract Settings
190(1)
Characterizing Mardia's Multivariate Pareto Distribution
191(2)
Bibliographic Notes
193(4)
Exercises
194(3)
Estimation in Conditionally Specified Models
197(32)
Introduction
197(1)
The Ubiquitous Norming Constant
197(1)
Maximum Likelihood
198(6)
Pseudolikelihood Involving Conditional Densities
204(3)
Marginal Likelihood
207(1)
An Efficiency Comparison
207(3)
Method of Moments Estimates
210(7)
Log-Linear Poisson Regression Estimates
217(2)
Bayesian Estimates
219(3)
Multivariate Examples
222(3)
Bibliographic Notes
225(4)
Exercises
225(4)
Marginal and Conditional Specification in General
229(26)
Introduction
229(1)
Specifying Bivariate Densities
229(6)
To Higher Dimensions with Care
235(1)
Conditional/Marginal Specification in k Dimensions
236(2)
Checking Uniqueness
238(2)
Checking Compatibility
240(6)
Overspecification
246(1)
Marginals and Conditionals in Specified Families
246(2)
Xi Given X(i)
248(1)
X(i) Given Xi
248(3)
The Case of X Given Y, Y Given Z, and Z Given X
251(1)
Logistic Regression Models
252(1)
Bibliographic Notes
253(2)
Exercises
253(2)
Conditional Survival Models
255(20)
Introduction
255(1)
Conditional Survival in the Bivariate Case
256(1)
Conditional Survival Functions in Parametric Families
257(4)
Examples of Distributions Characterized by Conditional Survival
261(3)
Multivariate Extensions
264(1)
Conditional Distributions
265(2)
Conditional Proportional Hazards
267(2)
Conditional Accelerated Failure
269(2)
An Alternative Specification Paradigm
271(3)
Bibliographic Notes
274(1)
Exercises
274(1)
Applications to Modeling Bivariate Extremes
275(18)
Introduction
275(1)
Univariate and Bivariate Gumbel Distributions
276(1)
Conditionally Specified Bivariate Gumbel Distributions
277(5)
Positive or Negative Correlation
282(1)
Density Contours
282(5)
Maximal Wind Speeds
287(3)
More Flexibility Needed
290(1)
Bibliographic Note
291(2)
Exercises
291(2)
Bayesian Analysis Using Conditionally Specified Models
293(44)
Introduction
293(1)
Motivation from a Normal Example
294(1)
Priors with Convenient Posteriors
295(4)
Conjugate Exponential Family Priors for Exponential Family Likelihoods
299(2)
Conditionally Specified Priors
301(3)
Normal Data
304(8)
Pareto Data
312(5)
Inverse Gaussian Data
317(1)
Ratios of Gamma Scale Parameters
318(4)
Comparison of Normal Means
322(5)
Regression
327(3)
The 2 x 2 Contingency Table
330(1)
Multinomial Data
331(1)
Change Point Problems
332(1)
Bivariate Normal
333(1)
No Free Lunch
334(1)
Bibliographic Notes
334(3)
Exercises
334(3)
Conditional Versus Simultaneous Equation Models
337(16)
Introduction
337(1)
Two Superficially Similar Models
338(1)
General CS and SE Models
339(1)
Linear Normal Models
340(4)
Nonlinear Normal Models
344(2)
Pareto CS and SE Models
346(3)
Discrete Models
349(1)
Higher-Dimensional Models
350(1)
Bibliographic Note
350(3)
Exercises
351(2)
Paella
353(18)
Introduction
353(1)
Diatomic Conditionals and Stop-Loss Transforms
353(4)
Failure Rates and Mean Residual Life Functions
357(1)
Hypothesis Testing
358(1)
Related Stochastic Processes
359(2)
Given E(X|Y = y) and FX(x)
361(1)
Near Compatibility with Given Parametric Families of Distributions
362(2)
Marginal Maps
364(1)
A Cautionary Tale
364(5)
Bibliographic Notes
369(2)
Exercises
369(2)
A Simulation 371(12)
A.1 Introduction
371(1)
A.2 The Rejection Method
372(2)
A.3 The Importance Sampling Method
374(3)
A.4 Application to Models with Conditionals in Exponential Families
377(1)
A.5 Other Conditionally Specified Models
378(1)
A.6 A Direct Approach Not Involving Rejection
379(1)
A.7 Bibliographic Notes
380(3)
Exercises
380(3)
B Notation Used in This Book 383(6)
References 389(10)
Author Index 399(4)
Subject Index 403

Supplemental Materials

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.

Rewards Program