Results 1 
8 of
8
ANCESTRAL GRAPH MARKOV MODELS
, 2002
"... This paper introduces a class of graphical independence models that is closed under marginalization and conditioning but that contains all DAG independence models. This class of graphs, called maximal ancestral graphs, has two attractive features: there is at most one edge between each pair of verti ..."
Abstract

Cited by 76 (18 self)
 Add to MetaCart
This paper introduces a class of graphical independence models that is closed under marginalization and conditioning but that contains all DAG independence models. This class of graphs, called maximal ancestral graphs, has two attractive features: there is at most one edge between each pair of vertices; every missing edge corresponds to an independence relation. These features lead to a simple parameterization of the corresponding set of distributions in the Gaussian case.
Chain Graph Models and their Causal Interpretations
 B
, 2001
"... Chain graphs are a natural generalization of directed acyclic graphs (DAGs) and undirected graphs. However, the apparent simplicity of chain graphs belies the subtlety of the conditional independence hypotheses that they represent. There are a number of simple and apparently plausible, but ultim ..."
Abstract

Cited by 48 (4 self)
 Add to MetaCart
Chain graphs are a natural generalization of directed acyclic graphs (DAGs) and undirected graphs. However, the apparent simplicity of chain graphs belies the subtlety of the conditional independence hypotheses that they represent. There are a number of simple and apparently plausible, but ultimately fallacious interpretations of chain graphs that are often invoked, implicitly or explicitly. These interpretations also lead to awed methods for applying background knowledge to model selection. We present a valid interpretation by showing how the distribution corresponding to a chain graph may be generated as the equilibrium distribution of dynamic models with feedback. These dynamic interpretations lead to a simple theory of intervention, extending the theory developed for DAGs. Finally, we contrast chain graph models under this interpretation with simultaneous equation models which have traditionally been used to model feedback in econometrics. Keywords: Causal model; cha...
Theory unification and graphical models in human categorization
"... Disparate, mutually incompatible theories of categorization are widespread in cognitive psychology. While there are various formal results connecting pairs of these theories, the primary research focus has been on particular empirical tests of people’s favorite theories. This chapter steps back from ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
Disparate, mutually incompatible theories of categorization are widespread in cognitive psychology. While there are various formal results connecting pairs of these theories, the primary research focus has been on particular empirical tests of people’s favorite theories. This chapter steps back from the question of which single theory (if any) is “right, ” and focuses
Systems Analysis by Graphtheoretic Techniques: Assessment of Institutional Linkages in the Agricultural Innovation System of
"... This paper develops a quantitative, graphtheoretic method for analysing systems of institutions. With an application to the agricultural innovation system of Azerbaijan, the method is illustrated in detail. An assessment of existing institutional linkages in the system suggests that efforts should ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
This paper develops a quantitative, graphtheoretic method for analysing systems of institutions. With an application to the agricultural innovation system of Azerbaijan, the method is illustrated in detail. An assessment of existing institutional linkages in the system suggests that efforts should be placed on the development of intermediary institutions to facilitate quick and effective flow of knowledge between the public and the private components of the system. Furthermore, significant accomplishments are yet to come in policymaking, research and education, and credit institutions.
Chain graphs \vhich are maxin1al ancestral graphs are recursive causal graphs
, 2001
"... In this note we prove that a chain graph is Markov equivalent to some DAG under marginalizing and conditioning if and only if it is Markov equivalent to a recursive causal graph. 1 ..."
Abstract
 Add to MetaCart
In this note we prove that a chain graph is Markov equivalent to some DAG under marginalizing and conditioning if and only if it is Markov equivalent to a recursive causal graph. 1
A Characterization of Markov Equivalence Classes for Directed Acyclic Graphs with Latent Variables
"... Different directed acyclic graphs (DAGs) may be Markov equivalent in the sense that they entail the same conditional independence relations among the observed variables. Meek (1995) characterizes Markov equivalence classes for DAGs (with no latent variables) by presenting a set of orientation rules ..."
Abstract
 Add to MetaCart
Different directed acyclic graphs (DAGs) may be Markov equivalent in the sense that they entail the same conditional independence relations among the observed variables. Meek (1995) characterizes Markov equivalence classes for DAGs (with no latent variables) by presenting a set of orientation rules that can correctly identify all arrow orientations shared by all DAGs in a Markov equivalence class, given a member of that class. For DAG models with latent variables, maximal ancestral graphs (MAGs) provide a neat representation that facilitates model search. Earlier work (Ali et al. 2005) has identified a set of orientation rules sufficient to construct all arrowheads common to a Markov equivalence class of MAGs. In this paper, we provide extra rules sufficient to construct all common tails as well. We end up with a set of orientation rules sound and complete for identifying commonalities across a Markov equivalence class of MAGs, which is particularly useful for causal inference. 1
An Assessment of Institutional Linkages
"... ISNAR, one of the 16 Future Harvest Centers supported by the Consultative Group on International Agricultural Research (CGIAR), seeks to contribute to the generation and use of knowledge that fosters sustainable and equitable agricultural development. ISNAR’s mission is to help bring about innovatio ..."
Abstract
 Add to MetaCart
ISNAR, one of the 16 Future Harvest Centers supported by the Consultative Group on International Agricultural Research (CGIAR), seeks to contribute to the generation and use of knowledge that fosters sustainable and equitable agricultural development. ISNAR’s mission is to help bring about innovation in agricultural research institutions to increase the contribution of research to agricultural development for the poor. ISNAR identifies and advances new arrangements that promote more effective generation of new knowledge. Emphasis on institutional innovation allows ISNAR to play a catalytic role in the change processes taking place in many developing countries. At the same time, it is strengthening its ability to play an important role in the new programs being developed by the CGIAR. The focus on institutional innovation puts ISNAR in a strategic position in a research community where national and international concerns are increasingly converging. ISNAR conducts its work on institutional innovation through the following six thematic areas of work: • Policies for institutional innovation for agricultural research • Linking research organizations and stakeholders in a changing context
Running title: Categorization and Graphical Models Contact information:
"... One natural representation of a category C is as a probability distribution (density) over the observed features. In this perspective, optimal categorization amounts to calculating the probability of that distribution given some novel observation. This paper focuses on probability distributions that ..."
Abstract
 Add to MetaCart
One natural representation of a category C is as a probability distribution (density) over the observed features. In this perspective, optimal categorization amounts to calculating the probability of that distribution given some novel observation. This paper focuses on probability distributions that can be represented using probabilistic graphical models, principally Bayesian networks and Markov random fields. The principal mathematical results in this paper prove that three prominent classes of psychological categorization models—those based on exemplars, prototypes, and causal models—are equivalent to optimal categorization in which the underlying probability distribution is restricted to (a subclass of) a particular type of probabilistic graphical model. That is, these psychological theories are all optimal categorizers, though they make different assumptions about the underlying structure of the world. The final section of this paper explores the significant theoretical and experimental implications of these equivalencies.