國立高雄大學圖資館 |

Vertex Processes in Social Networks.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Vertex Processes in Social Networks.
Author:	Almquist, Zack W.
Description:	163 p.
Notes:	Source: Dissertation Abstracts International, Volume: 74-10(E), Section: A.
Notes:	Adviser: Carter T. Butts.
Contained By:	Dissertation Abstracts International74-10A(E).
Subject:	Sociology, Theory and Methods.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3565375
ISBN:	9781303150883

Vertex Processes in Social Networks.
Almquist, Zack W.

Vertex Processes in Social Networks. - 163 p.

Source: Dissertation Abstracts International, Volume: 74-10(E), Section: A.

Thesis (Ph.D.)--University of California, Irvine, 2013.

Change in network structure and composition - either through dynamics and/or measurement error - has been a topic of extensive theoretical and methodological interest over the last two decades. Social networks represent a collection of social entities and relationships between these entities. Further, it is often convenient to represent a network as a mathematical object called a graph (G), which is comprised of two sets: a vertex set (e.g., organizations or individuals) and an edge set (e.g., friendship ties). Change in network structure can be induced by error or by measurement processes (e.g., error) or by social processes (i.e., intrinsic dynamics). This change in network structure can occur either through the vertex set (e.g. entry or exit of individuals into a network) or through the edge set (e.g., adding or deleting friendship ties).

ISBN: 9781303150883Subjects--Topical Terms:

212588
Sociology, Theory and Methods.

Vertex Processes in Social Networks.
LDR:05775nmm a2200337 4500 001 419280
005 20140520124009.5
008 140717s2013 ||||||||||||||||| ||eng d
020 $a 9781303150883
035 $a (MiAaPQ)AAI3565375
035 $a AAI3565375
040 $a MiAaPQ $c MiAaPQ
100 1 $a Almquist, Zack W. $3 660311
245 1 0 $a Vertex Processes in Social Networks.
300 $a 163 p.
500 $a Source: Dissertation Abstracts International, Volume: 74-10(E), Section: A.
500 $a Adviser: Carter T. Butts.
502 $a Thesis (Ph.D.)--University of California, Irvine, 2013.
520 $a Change in network structure and composition - either through dynamics and/or measurement error - has been a topic of extensive theoretical and methodological interest over the last two decades. Social networks represent a collection of social entities and relationships between these entities. Further, it is often convenient to represent a network as a mathematical object called a graph (G), which is comprised of two sets: a vertex set (e.g., organizations or individuals) and an edge set (e.g., friendship ties). Change in network structure can be induced by error or by measurement processes (e.g., error) or by social processes (i.e., intrinsic dynamics). This change in network structure can occur either through the vertex set (e.g. entry or exit of individuals into a network) or through the edge set (e.g., adding or deleting friendship ties).
520 $a The systematic errors that are induced by a combination of human memory limitations and common survey design and implementation have long been studied in the context of egocentric networks. Despite this, little if any work exists in the area of random error analysis on these same networks; this chapter offers a perspective on the effects of random errors on egonet analysis, as well as the effects of using egonet measures as independent predictors in linear models. Notice that we may treat these errors as a vertex process (the inclusion or exclusion of entities within the network and the resulting effect on structure). We explore the effects of false-positive and false-negative error in egocentric networks on both standard network measures and on linear models through simulation analysis on a ground truth egocentric network sample based on Facebook-friendships. Results show that 5-20% error rates, which are consistent with error rates known to occur in egocentric network data, can cause serious misestimation of network properties and regression parameters.
520 $a Network dynamics may be viewed as a process of change in the edge structure of a network, in the vertex set on which edges are defined, or in both simultaneously. Though early studies of such processes were primarily descriptive, recent work on this topic has increasingly turned to formal statistical models. While showing great promise, many of these modern dynamic models are computationally intensive and scale very poorly in the size of the network under study and/or the number of time points considered. Likewise, currently employed models focus on edge dynamics, with little support for endogenously changing vertex sets. Here, we show how an existing approach based on logistic network regression can be extended to serve as highly scalable framework for modeling large networks with dynamic vertex sets. We place this approach within a general dynamic exponential family (ERGM) context, clarifying the assumptions underlying the framework (and providing a clear path for extensions), and show how model assessment methods for cross-sectional networks can be extended to the dynamic case. Finally, we illustrate this approach on a classic data set involving interactions among windsurfers on a California beach.
520 $a Methods for dynamic network analysis have greatly advanced in recent decade. This chapter extends current methods of dynamic network logistic regression (a member of the Temporal Exponential Random Graph Models) to network-panel data which contains missing data in the edge set and vertex set. We begin by reviewing the non-missing data form of dynamic network logistic regression. We then layout a missing data framework akin to that of Little and Rubin (2002) and Handcock and Gile (2010). We discuss different methods for dealing with missing data, including multiple imputation. Working within this likelihood based framework we will derive a Metropolis-Hastings algorithm for imputing the missing values based on the full temporal conditionals of the (updated) logistic model and the observed data. Further, we note computational complexity of the multiple imputation method in the DNR case and propose a simpler model based approach that takes advantage of the assumptions of DNR. We dub this method the complete-case method. Finally, we derive the likelihood and perform simulation analysis to two real world datasets.
520 $a All three chapters put-together add significantly to our understanding of how vertex processes affect network models and measurement. Here we have explored the effect of change in vertex composition as first an error process (addition or deletion of vertices on statistic network), as a dynamic process (entry and exit of vertices over time), and finally as both an error and dynamic process (i.e., network dynamics with missing data). The results that clearly show the importance of modeling or measuring the vertex set in many applications of interest to social scientists.
590 $a School code: 0030.
650 4 $a Sociology, Theory and Methods. $3 212588
650 4 $a Statistics. $3 182057
650 4 $a Sociology, Demography. $3 227615
690 $a 0344
690 $a 0463
690 $a 0938
710 2 $a University of California, Irvine. $b Sociology - Ph.D.. $3 660312
773 0 $t Dissertation Abstracts International $g 74-10A(E).
790 $a 0030
791 $a Ph.D.
792 $a 2013
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3565375