Analysis of the threshold values of semantic structure analysis in identifying causal relationships

Jiunn I. Shieh, Hsin-Hung Wu, Hsiang Chuan Liu

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

This study uses a semantic structure analysis (SSA) method to construct the causal relationships among the criteria from survey data. The literatures provide a predetermined threshold value when the SSA is applied without explanation, but we use a Monte Carlo simulation based on the raw data to determine the threshold values with the significant levels of 0.05 and 0.10 for constructing the causal relationships. The results show that the causal relationships among the criteria using the suggested threshold value are too complicated, while the causal relationships by the simulated threshold values are relatively easy to be understood and used practically.

Original languageEnglish
Pages (from-to)1543-1551
Number of pages9
JournalCommunications in Statistics: Simulation and Computation
Volume43
Issue number7
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

Threshold Value
Semantics
Survey Data
Monte Carlo Simulation
Relationships
Monte Carlo simulation

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Modelling and Simulation

Cite this

@article{e4547d33658343ea8f44ddae002c13e3,
title = "Analysis of the threshold values of semantic structure analysis in identifying causal relationships",
abstract = "This study uses a semantic structure analysis (SSA) method to construct the causal relationships among the criteria from survey data. The literatures provide a predetermined threshold value when the SSA is applied without explanation, but we use a Monte Carlo simulation based on the raw data to determine the threshold values with the significant levels of 0.05 and 0.10 for constructing the causal relationships. The results show that the causal relationships among the criteria using the suggested threshold value are too complicated, while the causal relationships by the simulated threshold values are relatively easy to be understood and used practically.",
author = "Shieh, {Jiunn I.} and Hsin-Hung Wu and Liu, {Hsiang Chuan}",
year = "2014",
month = "1",
day = "1",
doi = "10.1080/03610918.2012.735322",
language = "English",
volume = "43",
pages = "1543--1551",
journal = "Communications in Statistics Part B: Simulation and Computation",
issn = "0361-0918",
publisher = "Taylor and Francis Ltd.",
number = "7",

}

Analysis of the threshold values of semantic structure analysis in identifying causal relationships. / Shieh, Jiunn I.; Wu, Hsin-Hung; Liu, Hsiang Chuan.

In: Communications in Statistics: Simulation and Computation, Vol. 43, No. 7, 01.01.2014, p. 1543-1551.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Analysis of the threshold values of semantic structure analysis in identifying causal relationships

AU - Shieh, Jiunn I.

AU - Wu, Hsin-Hung

AU - Liu, Hsiang Chuan

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This study uses a semantic structure analysis (SSA) method to construct the causal relationships among the criteria from survey data. The literatures provide a predetermined threshold value when the SSA is applied without explanation, but we use a Monte Carlo simulation based on the raw data to determine the threshold values with the significant levels of 0.05 and 0.10 for constructing the causal relationships. The results show that the causal relationships among the criteria using the suggested threshold value are too complicated, while the causal relationships by the simulated threshold values are relatively easy to be understood and used practically.

AB - This study uses a semantic structure analysis (SSA) method to construct the causal relationships among the criteria from survey data. The literatures provide a predetermined threshold value when the SSA is applied without explanation, but we use a Monte Carlo simulation based on the raw data to determine the threshold values with the significant levels of 0.05 and 0.10 for constructing the causal relationships. The results show that the causal relationships among the criteria using the suggested threshold value are too complicated, while the causal relationships by the simulated threshold values are relatively easy to be understood and used practically.

UR - http://www.scopus.com/inward/record.url?scp=84893938834&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893938834&partnerID=8YFLogxK

U2 - 10.1080/03610918.2012.735322

DO - 10.1080/03610918.2012.735322

M3 - Article

AN - SCOPUS:84893938834

VL - 43

SP - 1543

EP - 1551

JO - Communications in Statistics Part B: Simulation and Computation

JF - Communications in Statistics Part B: Simulation and Computation

SN - 0361-0918

IS - 7

ER -