Support Vector Regression and Beta Distribution for the Modeling of Incumbent Party vote for Presidential Elections
The aim of our study is to model and predict, rather than explain presidential election results, using selected quarterly macroeconomic indicators, say, gross national product, consumer price index, unemployment rate and gross national product from 1994-2017.Particularly, we seek to provide predict...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Applied Mathematics & Information Sciences
2021
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/20.500.12493/475 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1800403070434148352 |
|---|---|
| author | R., Kikawa M, N. Ngungu D., Ntirampeba A., Ssematimba |
| author_facet | R., Kikawa M, N. Ngungu D., Ntirampeba A., Ssematimba |
| author_sort | R., Kikawa |
| collection | KAB-DR |
| description | The aim of our study is to model and predict, rather than explain presidential election results, using selected quarterly
macroeconomic indicators, say, gross national product, consumer price index, unemployment rate and gross national product from 1994-2017.Particularly, we seek to provide predictions of presidential winner prior to the elections based on the beta distribution and the support vector regression (SVR) as prediction models.Two models are primarily built based on beta distribution and SVR. Due to the forecasting aspect, model performance is focused on mainly one goodness-of-fit measure, that is, the prediction error rather than the squared correlation coefficient R2 as it makes little sense in a practical regression perspective. The best model is the one with the least mean square error (MSE). In this effect it turns out that, the SVR with kernel type encapsulated postscript eps radial has a mean square error of 0.006 on the test set is a better model as compared to the beta distribution model with a mean square error of 1.216. An accurate solution to prediction of presidential vote elections via SVR analysis is therefore proposed. |
| format | Article |
| id | oai:idr.kab.ac.ug:20.500.12493-475 |
| institution | KAB-DR |
| language | English |
| publishDate | 2021 |
| publisher | Applied Mathematics & Information Sciences |
| record_format | dspace |
| spelling | oai:idr.kab.ac.ug:20.500.12493-4752024-01-17T04:45:01Z Support Vector Regression and Beta Distribution for the Modeling of Incumbent Party vote for Presidential Elections R., Kikawa M, N. Ngungu D., Ntirampeba A., Ssematimba Proportions,Time series data, Predictions, Regression, Support Vector Machine The aim of our study is to model and predict, rather than explain presidential election results, using selected quarterly macroeconomic indicators, say, gross national product, consumer price index, unemployment rate and gross national product from 1994-2017.Particularly, we seek to provide predictions of presidential winner prior to the elections based on the beta distribution and the support vector regression (SVR) as prediction models.Two models are primarily built based on beta distribution and SVR. Due to the forecasting aspect, model performance is focused on mainly one goodness-of-fit measure, that is, the prediction error rather than the squared correlation coefficient R2 as it makes little sense in a practical regression perspective. The best model is the one with the least mean square error (MSE). In this effect it turns out that, the SVR with kernel type encapsulated postscript eps radial has a mean square error of 0.006 on the test set is a better model as compared to the beta distribution model with a mean square error of 1.216. An accurate solution to prediction of presidential vote elections via SVR analysis is therefore proposed. Kabale University 2021-01-27T06:53:15Z 2021-01-27T06:53:15Z 2020 Article http://hdl.handle.net/20.500.12493/475 en application/pdf Applied Mathematics & Information Sciences |
| spellingShingle | Proportions,Time series data, Predictions, Regression, Support Vector Machine R., Kikawa M, N. Ngungu D., Ntirampeba A., Ssematimba Support Vector Regression and Beta Distribution for the Modeling of Incumbent Party vote for Presidential Elections |
| title | Support Vector Regression and Beta Distribution for the Modeling of Incumbent Party vote for Presidential Elections |
| title_full | Support Vector Regression and Beta Distribution for the Modeling of Incumbent Party vote for Presidential Elections |
| title_fullStr | Support Vector Regression and Beta Distribution for the Modeling of Incumbent Party vote for Presidential Elections |
| title_full_unstemmed | Support Vector Regression and Beta Distribution for the Modeling of Incumbent Party vote for Presidential Elections |
| title_short | Support Vector Regression and Beta Distribution for the Modeling of Incumbent Party vote for Presidential Elections |
| title_sort | support vector regression and beta distribution for the modeling of incumbent party vote for presidential elections |
| topic | Proportions,Time series data, Predictions, Regression, Support Vector Machine |
| url | http://hdl.handle.net/20.500.12493/475 |
| work_keys_str_mv | AT rkikawa supportvectorregressionandbetadistributionforthemodelingofincumbentpartyvoteforpresidentialelections AT mnngungu supportvectorregressionandbetadistributionforthemodelingofincumbentpartyvoteforpresidentialelections AT dntirampeba supportvectorregressionandbetadistributionforthemodelingofincumbentpartyvoteforpresidentialelections AT assematimba supportvectorregressionandbetadistributionforthemodelingofincumbentpartyvoteforpresidentialelections |