Symbolic analysis of indicator time series by quantitative sequence alignment

Authors:
Takuya Yamano;Kodai Sato;Taisei Kaizoji;Jan-Michael Rost;Lukáš Pichl
Affiliations:
International Christian University, Osawa 3-10-2, Mitaka, Tokyo, 181-8585, Japan;International Christian University, Osawa 3-10-2, Mitaka, Tokyo, 181-8585, Japan;International Christian University, Osawa 3-10-2, Mitaka, Tokyo, 181-8585, Japan;Max Planck Institute for the Physics of Complex Systems, Nöthnitzer Straíe 38, Dresden, 01187, Germany;Max Planck Institute for the Physics of Complex Systems, Nöthnitzer Straíe 38, Dresden, 01187, Germany
Venue:
Computational Statistics & Data Analysis
Year:
2008

Citing 3
Cited 1

Random number generators: good ones are hard to find

Communications of the ACM
Algorithms for the Longest Common Subsequence Problem

Journal of the ACM (JACM)
Optimal Multi-Scale Time Series Decomposition for Financial Forecasting using Wavelet Thresholding Techniques

RSFDGrC '99 Proceedings of the 7th International Workshop on New Directions in Rough Sets, Data Mining, and Granular-Soft Computing

Capability and limitation of financial time-series data prediction using symbol string quantization

Proceedings of the 2009 International Conference on Hybrid Information Technology

Quantified Score

Hi-index	0.03

Visualization

Abstract

Symbolic analysis of economic indicators and derived time series offers an advantage of transferring quantitative values into qualitative notions by indexing intervals of numerical data with symbols. While differences in the numerical indicators are routinely measured by subtraction, differences in the symbolic indicators can be compared via more procedural quantitative-scoring schemes, the complexity of which depends on the alphabet size. In effect, the similarity of symbolic data sequence becomes a subtle measure. Upon motivating principles of symbolic analysis, our analysis illustrates how the optimized numerical scoring for alignment schemes may reveal functional and causal relations among the indicator data. The approach of symbolic analysis is particularly suitable for data processing in economics, in which partitioning of resources, competence, information access, or knowledge representation is common by the methodological design.