Title | ||
---|---|---|
Finding the clusters with potential value in financial time series based on agglomerative hierarchical clustering |
Abstract | ||
---|---|---|
It is interesting to find the clustering with potential value in financial time series. In this paper, we focus on this topic. The owned features of the clusters with potential value are provided firstly. Then, the agglomerative hierarchical clustering (AHC) is used to find those clusters automatically. There are two innovations in this paper. The first one is that the features of the clusters with potential value are embedded into the process of AHC, which reduces the time cost of clustering process. The second one is that we propose two indicators, whole similarity and trend similarity, to measure the persistence of the cluster. The experiment on ten time segments shows the obtained clusters is effective, in which both the whole similarity and the trend similarity on training data are markedly higher than that of randomized clustering. In addition, the persistence of these clusters on test data is also better that the result of randomized guess. We think that the strategy provided in this paper is helpful to find for the clustering with potential value. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1109/ICCSE.2016.7581558 | 2016 11th International Conference on Computer Science & Education (ICCSE) |
Keywords | Field | DocType |
financial time series,agglomerative hierarchical clustering,persistence of cluster | Data mining,CURE data clustering algorithm,Computer science,Artificial intelligence,Cluster analysis,Single-linkage clustering,Hierarchical clustering,Complete-linkage clustering,Correlation clustering,Determining the number of clusters in a data set,Brown clustering,Finance,Machine learning | Conference |
ISSN | ISBN | Citations |
2471-6146 | 978-1-5090-2219-9 | 0 |
PageRank | References | Authors |
0.34 | 2 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shi Yang You | 1 | 0 | 0.34 |
Yu Dan Wang | 2 | 0 | 0.34 |
Linkai Luo | 3 | 163 | 14.00 |
Hong Peng | 4 | 14 | 10.33 |