Muutke küpsiste eelistusi

Inferential Network Analysis [Pehme köide]

4.00/5 (2 hinnangut Goodreads-ist)
(The Ohio State University), (The Ohio State University), (Pennsylvania State University)
  • Formaat: Paperback / softback, 314 pages, kõrgus x laius x paksus: 226x151x18 mm, kaal: 470 g
  • Sari: Analytical Methods for Social Research
  • Ilmumisaeg: 19-Nov-2020
  • Kirjastus: Cambridge University Press
  • ISBN-10: 1316610853
  • ISBN-13: 9781316610855
Teised raamatud teemal:
Inferential Network AnalysisSuurem pilt
  • Formaat: Paperback / softback, 314 pages, kõrgus x laius x paksus: 226x151x18 mm, kaal: 470 g
  • Sari: Analytical Methods for Social Research
  • Ilmumisaeg: 19-Nov-2020
  • Kirjastus: Cambridge University Press
  • ISBN-10: 1316610853
  • ISBN-13: 9781316610855
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781316610855.html
Märksõnad:
This unique textbook provides an introduction to statistical inference with network data. The authors present a self-contained derivation and mathematical formulation of methods, review examples, and real-world applications, as well as provide data and code in the R environment that can be customised. Inferential network analysis transcends fields, and examples from across the social sciences are discussed (from management to electoral politics), which can be adapted and applied to a panorama of research. From scholars to undergraduates, spanning the social, mathematical, computational and physical sciences, readers will be introduced to inferential network models and their extensions. The exponential random graph model and latent space network model are paid particular attention and, fundamentally, the reader is given the tools to independently conduct their own analyses.

Arvustused

'The family of exponential random graph models have advanced with a number of extensions in recent years, many of them developed by the present authors. Encapsulating these advances with other methods of inferential analysis in a single reference that combines essential theory with hands-on examples makes this book a must-have for network modeling practitioners who want to use these powerful tools.' Peter Mucha, UNC Chapel Hill

Muu info

Pioneering introduction of unprecedented breadth and scope to inferential and statistical methods for network analysis.
List of Figures
viii
List of Tables
xiii
Notation and Acronyms xiv
Preface xix
Acknowledgments xxiii
PART I DEPENDENCE AND INTERDEPENDENCE
1 Promises and Pitfalls of Inferential Network Analysis
3(14)
1.1 A Basis for Considering Networks
4(4)
1.2 Networks and Complex Statistical Dependence
8(6)
1.3 Methods Covered in This Book
14(3)
2 Detecting and Adjusting for Network Dependencies
17(18)
2.1 Detecting Dependencies: Conditional Uniform Graph Tests
19(7)
2.2 The Quadratic Assignment Procedure (QAP)
26(5)
2.3 Wrapping Up
31(1)
2.4 Self-Study Problems
32(3)
PART II THE FAMILY OF EXPONENTIAL RANDOM GRAPH MODELS (ERGMS)
3 The Basic ERGM
35(32)
3.1 Introduction
35(5)
3.2 The Exponential Random Graph Model (ERGM)
40(2)
3.3 ERGM Specification: A Brief Introduction
42(9)
3.4 Model Fit
51(5)
3.5 Interpretation
56(8)
3.6 Limitations
64(1)
3.7 Wrapping Up
65(1)
3.8 Self-Study Problems
66(1)
4 ERGM Specification
67(31)
4.1 Starting with Theory
68(2)
4.2 Exogenous Covariate Effects
70(5)
4.3 Endogenous Network Effects
75(14)
4.4 Creating New Statistics
89(1)
4.5 Bipartite ERGMs
90(6)
4.6 Wrapping Up
96(1)
4.7 Self-Study Problems
97(1)
5 Estimation and Degeneracy
98(18)
5.1 Methods for Estimating ERGM
98(5)
5.2 Problem of Degeneracy
103(3)
5.3 Adjusting Specifications to Correct Degeneracy and Improve Model Fit
106(8)
5.4 Other Estimation Methods for ERGMs
114(1)
5.5 Wrapping Up
115(1)
5.6 Self-Study Problems
115(1)
6 ERG Type Models for Longitudinally Observed Networks
116(32)
6.1 Introduction
116(1)
6.2 Data Considerations
117(5)
6.3 The Temporal Exponential Random Graph Model (TERGM)
122(2)
6.4 TERGM Specification
124(7)
6.5 To Pool or Not to Pool? Temporal Stability of Effects
131(3)
6.6 Estimation
134(4)
6.7 The Stochastic Actor-Oriented Model (SAOM)
138(8)
6.8 Wrapping Up
146(1)
6.9 Self-Study Problems
146(2)
7 Valued-Edge ERGMs: The Generalized ERGM (GERGM)
148(19)
7.1 GERGM Definition
150(3)
7.2 Specifying Processes on Weighted Networks
153(1)
7.3 Avoiding Degeneracy in the GERGM
154(2)
7.4 Parameter Estimation
156(3)
7.5 Applications in the Literature
159(4)
7.6 Wrapping Up
163(1)
7.7 Self-Study Problems
163(4)
PART III LATENT SPACE NETWORK MODELS
8 The Basic Latent Space Model
167(53)
8.1 Introduction
167(1)
8.2 Motivation: Theoretical and Mathematical Perspective
168(3)
8.3 The Euclidean Latent Space Model
171(6)
8.4 Model Convergence
177(8)
8.5 Model Fit
185(4)
8.6 Model Specification
189(23)
8.7 Interpretation of Latent Space Models
212(5)
8.8 Strengths, Assumptions, and Limitations of the Latent Space Model
217(1)
8.9 Wrapping Up
218(1)
8.10 Self-Study Problems
218(2)
9 Identification, Estimation, and Interpretation of the Latent Space Model
220(16)
9.1 Parameter Identification
221(8)
9.2 Identification: Some Solutions
229(2)
9.3 Interpreting the Latent Space
231(1)
9.4 The Problem with Isolates
232(1)
9.5 Estimation
233(2)
9.6 Wrapping Up
235(1)
10 Extending the Latent Space Model
236(36)
10.1 Introduction
236(12)
10.2 Valued-Edge Networks
248(7)
10.3 Cluster Models
255(6)
10.4 Random Effects Models
261(4)
10.5 The Additive and Multiplicative Effects Latent Factor Model (LFM)
265(4)
10.6 Other Extensions
269(1)
10.7 Wrapping Up
270(1)
10.8 Self-Study Problems
271(1)
References 272(16)
Index 288
Skyler J. Cranmer is the Carter Phillips and Sue Henry Professor of Political Science at The Ohio State University. Bruce A. Desmarais is the DeGrandis-McCourtney Early Career Professor in Political Science at Penn State University. Jason William Morgan is the Vice President for Behavioural Intelligence: Aware, and visiting scholar in Political Science at The Ohio State University.