Abstract
The objective of this paper is to continue the study of inductive approaches to determine causality in comparative politics by a fuzzy approach and then a classical statistical method. A method proposed by Guiasu [1] was utilized in [2] whereas the Jeffrey's Rule as described in [1] was employed in [3]. In this paper, the Bayes' Rule as described in [1] is used to combine the democracy values of Freedom House, Vanhanen, and Polity for 27 countries in Eastern Europe. Also, the Dempster-Schafer method is also applied to measure the democracy values for these 27 countries. A data set of eight factors hypothesized to be important is used to demonstrate consolidation. These factors are: H1 (stateness), H2 (corruption), H3 (privatization), H4 (multi-party system), H5 (usable - reformed - public administration), H6 (decentralization), H7 (rule of law - autonomous judiciary), and H8 (avoiding a powerful chief executive). In [1], we use a model proposed by Guiasu that describes the process of reaching a verdict by probabilistic weighting of available evidence. We used the model to determine a numerical relationship between the H-factors. The classical rules from decision theory proposed by Hooper, Dempster, Bayes, and Jeffrey are special cases of Guiasu's weighting process. For each H-factor, two sets of median democracy values for all 27 countries in Eastern Europe were obtained - one by the Freedom House, Vanhanen, and Polity method and the other by the Dempster-Schafer method. Subsequently, an extensive non-parametric statistical analysis will be performed on scores of these 8 H-factors to investigate the consistency of the two scoring methods.
| Original language | English |
|---|---|
| Title of host publication | NAFIPS 2007: 2007 Annual Meeting of the North American Fuzzy Information Processing Society |
| Pages | 164-167 |
| Number of pages | 4 |
| DOIs | |
| State | Published - 2007 |
| Event | NAFIPS 2007: 2007 Annual Meeting of the North American Fuzzy Information Processing Society - San Diego, CA, United States Duration: Jun 24 2007 → Jun 27 2007 |
Other
| Other | NAFIPS 2007: 2007 Annual Meeting of the North American Fuzzy Information Processing Society |
|---|---|
| Country | United States |
| City | San Diego, CA |
| Period | 6/24/07 → 6/27/07 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Media Technology
Cite this
Combination of democracy values using Hooper's rule. / Cheng, Shih-Chuan; Grieser, Adam D.; Clark, Terry D.; Mordeson, John N.
NAFIPS 2007: 2007 Annual Meeting of the North American Fuzzy Information Processing Society. 2007. p. 164-167 4271053.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Combination of democracy values using Hooper's rule
AU - Cheng, Shih-Chuan
AU - Grieser, Adam D.
AU - Clark, Terry D.
AU - Mordeson, John N.
PY - 2007
Y1 - 2007
N2 - The objective of this paper is to continue the study of inductive approaches to determine causality in comparative politics by a fuzzy approach and then a classical statistical method. A method proposed by Guiasu [1] was utilized in [2] whereas the Jeffrey's Rule as described in [1] was employed in [3]. In this paper, the Bayes' Rule as described in [1] is used to combine the democracy values of Freedom House, Vanhanen, and Polity for 27 countries in Eastern Europe. Also, the Dempster-Schafer method is also applied to measure the democracy values for these 27 countries. A data set of eight factors hypothesized to be important is used to demonstrate consolidation. These factors are: H1 (stateness), H2 (corruption), H3 (privatization), H4 (multi-party system), H5 (usable - reformed - public administration), H6 (decentralization), H7 (rule of law - autonomous judiciary), and H8 (avoiding a powerful chief executive). In [1], we use a model proposed by Guiasu that describes the process of reaching a verdict by probabilistic weighting of available evidence. We used the model to determine a numerical relationship between the H-factors. The classical rules from decision theory proposed by Hooper, Dempster, Bayes, and Jeffrey are special cases of Guiasu's weighting process. For each H-factor, two sets of median democracy values for all 27 countries in Eastern Europe were obtained - one by the Freedom House, Vanhanen, and Polity method and the other by the Dempster-Schafer method. Subsequently, an extensive non-parametric statistical analysis will be performed on scores of these 8 H-factors to investigate the consistency of the two scoring methods.
AB - The objective of this paper is to continue the study of inductive approaches to determine causality in comparative politics by a fuzzy approach and then a classical statistical method. A method proposed by Guiasu [1] was utilized in [2] whereas the Jeffrey's Rule as described in [1] was employed in [3]. In this paper, the Bayes' Rule as described in [1] is used to combine the democracy values of Freedom House, Vanhanen, and Polity for 27 countries in Eastern Europe. Also, the Dempster-Schafer method is also applied to measure the democracy values for these 27 countries. A data set of eight factors hypothesized to be important is used to demonstrate consolidation. These factors are: H1 (stateness), H2 (corruption), H3 (privatization), H4 (multi-party system), H5 (usable - reformed - public administration), H6 (decentralization), H7 (rule of law - autonomous judiciary), and H8 (avoiding a powerful chief executive). In [1], we use a model proposed by Guiasu that describes the process of reaching a verdict by probabilistic weighting of available evidence. We used the model to determine a numerical relationship between the H-factors. The classical rules from decision theory proposed by Hooper, Dempster, Bayes, and Jeffrey are special cases of Guiasu's weighting process. For each H-factor, two sets of median democracy values for all 27 countries in Eastern Europe were obtained - one by the Freedom House, Vanhanen, and Polity method and the other by the Dempster-Schafer method. Subsequently, an extensive non-parametric statistical analysis will be performed on scores of these 8 H-factors to investigate the consistency of the two scoring methods.
UR - http://www.scopus.com/inward/record.url?scp=35148896348&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=35148896348&partnerID=8YFLogxK
U2 - 10.1109/NAFIPS.2007.383830
DO - 10.1109/NAFIPS.2007.383830
M3 - Conference contribution
AN - SCOPUS:35148896348
SN - 1424412145
SN - 9781424412143
SP - 164
EP - 167
BT - NAFIPS 2007: 2007 Annual Meeting of the North American Fuzzy Information Processing Society
ER -