DBHT
PortfolioOptimisers.UniqueRoot — Type
struct UniqueRoot <: DBHTRootMethodJoins every root candidate of the planar hierarchy under one root.
When the clique hierarchy has one root candidate, that 3-clique is the root. When it has several, CliqueRoot adds one synthetic clique and makes it the parent of every candidate. That clique is the imaginary 3-clique of [126]. Its interior is the whole graph, so the hierarchy has one maximal element.
Related
References
- [126] W.-M. Song, T. Di Matteo and T. Aste. Nested hierarchies in planar graphs. Discrete Applied Mathematics 159, 2135–2146 (2011). Definition 4.1 and Corollary 4.2.
PortfolioOptimisers.EqualRoot — Type
struct EqualRoot <: DBHTRootMethodJoins the root candidates of the planar hierarchy to each other, and adds no root clique.
The hierarchy keeps every root candidate, and it puts no candidate above another. CliqueRoot joins two root candidates when they share two vertices.
[126] leaves the hierarchy a forest of one tree for each candidate, and states no rule that joins them. The rule comes from the MATLAB code that accompanies [127], and no published work states it.
Related
References
- [126] W.-M. Song, T. Di Matteo and T. Aste. Nested hierarchies in planar graphs. Discrete Applied Mathematics 159, 2135–2146 (2011). Corollary 3.6. The original source of this formulation was not found. The formulation follows this work and is not checked against the original.
PortfolioOptimisers.DBHT — Type
struct DBHT{__T_sim<:AbstractNonNegativeSimilarityMatrixAlgorithm, __T_root} <: AbstractHierarchicalClusteringAlgorithmClusters assets by the bubble hierarchy of a triangulated maximally filtered graph.
DBHT selects the Direct Bubble Hierarchical Tree method of [127] as the algorithm of a ClustersEstimator. clusterise then builds the dendrogram with DBHTs.
[127] builds the planar maximally filtered graph of [128], which inserts edges one at a time. This type builds the triangulated maximally filtered graph of [129] with PMFG_T2s instead, as the MATLAB code that accompanies [127] does. Both are maximal planar graphs, so the bubble hierarchy of [127] applies to either.
Fields
sim: Similarity matrix algorithm that maps the distance matrix to the weights of the PMFG. The PMFG cannot take a negative weight, so the field takes the non-negative family alone, and the constructor refusesAngularSimilarity.
root: The rule that makes one root of the clique hierarchy when the planar graph gives more than one root candidate, a member ofDBHTRootMethod.
Constructors
DBHT(; sim::AbstractNonNegativeSimilarityMatrixAlgorithm = MaximumDistanceSimilarity(), root::DBHTRootMethod = UniqueRoot()) -> DBHTKeywords correspond to the struct's fields.
Examples
julia> DBHT()DBHT sim ┼ MaximumDistanceSimilarity() root ┴ UniqueRoot()Related
AbstractHierarchicalClusteringAlgorithmAbstractNonNegativeSimilarityMatrixAlgorithmAbstractSimilarityMatrixAlgorithmDBHTRootMethodMaximumDistanceSimilarityExponentialSimilarityGeneralExponentialSimilarityUniqueRootEqualRoot
References
- [127] W.-M. Song, T. Di Matteo and T. Aste. Hierarchical information clustering by means of topologically embedded graphs. PloS one 7, e31929 (2012). Section "Methods", Equations 1-5.
- [128] M. Tumminello, T. Aste, T. Di Matteo and R. N. Mantegna. A tool for filtering information in complex systems. Proceedings of the National Academy of Sciences 102, 10421–10426 (2005). For the planar maximally filtered graph of the original DBHT method.
- [129] G. P. Massara, T. Di Matteo and T. Aste. Network Filtering for Big Data: Triangulated Maximally Filtered Graph. Journal of Complex Networks 5, 161–178 (2016). Section III.A, Algorithm 1, for the graph that this type builds.
References
- [126]
- W.-M. Song, T. D. Matteo and T. Aste. Nested hierarchies in planar graphs. Discrete Applied Mathematics 159, 2135–2146 (2011).
- [127]
- W.-M. Song, T. Di Matteo and T. Aste. Hierarchical information clustering by means of topologically embedded graphs. PloS one 7, e31929 (2012).
- [128]
- M. Tumminello, T. Aste, T. Di Matteo and R. N. Mantegna. A tool for filtering information in complex systems. Proceedings of the National Academy of Sciences 102, 10421–10426 (2005).
- [129]
- G. P. Massara, T. Di Matteo and T. Aste. Network Filtering for Big Data: Triangulated Maximally Filtered Graph. Journal of Complex Networks 5, 161–178 (2016), arXiv:https://academic.oup.com/comnet/article-pdf/5/2/161/13794756/cnw015.pdf.