Subjective measurements of persistence of time series

Title: Subjective measurements of persistence of time series
Creator: Poswayo, Sihle
Subject: Time-series analysis
Subject: Space and time Time -- Philosophy
Date Issued: 2018
Date: 2018
Type: Thesis
Type: Masters
Type: MSc
Identifier: http://hdl.handle.net/10948/34372
Identifier: vital:33361
Description: In this paper we suggest the use of subjective judgements to measure persistence in time series by comparing pairs of graphs with different Hurst exponents. The group of respondents consisted of 40 volunteers who were asked to identify the more jagged out of two graphs presented to them (that is, less persistent). The respondents were approached as a group and requested to work independently in the completion of ques- tionnaires administered to them. The respondents were supervised by the researchers. The graphs were simulated using time series package of Mathematica R [26]. The re- sponses were processed using an algorithm based on the Thurstone-Mosteller model for paired comparisons [29]. The results of the analysis show that the human eye is capable of distinguishing graphs of time series with Hurst exponent difference as small as only 0.02.
Format: x, 83 leaves
Format: pdf
Publisher: Nelson Mandela University
Publisher: Faculty of Science
Language: English
Rights: Nelson Mandela University

Collections

NMU School of Computer Science, Mathematics, Physics and Statistics

		Thumbnail	File	Description	Size	Format
View Details Download			SOURCE1	Sihle Poswayo.pdf	766 KB	Adobe Acrobat PDF	View Details Download