Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach

Jeremiah Kiingati; Samuel Mwalili; Anthony Waititu

doi:doi:10.11648/j.ijdsa.20200604.13

| Peer-Reviewed

Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach

Jeremiah Kiingati, Samuel Mwalili, Anthony Waititu

Published in International Journal of Data Science and Analysis (Volume 6, Issue 4)

Received: 17 August 2020 Accepted: 5 September 2020 Published: 19 September 2020

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

In this paper we apply sequential Bayesian approach to compare the outcome of the presidential polls in Kenya. We use the previous polls to form the prior for the current polls. Even though several authors have used non-Bayesian models for countrywide polling data to forecast the outcome of the presidential race we propose a Bayesian approach in this case. As such the question of how to treat the previous and current pre-election polls data is inevitable. Some researchers consider only the most recent poll others Combine all previous polls up the present time and treat it as a single sample, weighting only by sample size, while others Combine all previous polls but adjust the sample size according to a weight function depending on the day the poll is taken. In this paper we apply a sequential Bayesian model (as an advancement of the latter which is time sensitive) where the previous measure is used as the prior of the current measure. Our concern is to model the proportion of votes between two candidates, incumbent and challenger. A Bayesian model of our binomial variable of interest will be applied sequentially to the Kenya opinion poll data sets in order to arrive at a posterior probability statement. The simulation results show that the eventual winner must lead consistently and constantly in at least 60% of the opinions polls. In addition, a candidate demonstrating high variability is more likely to lose the polls.

Published in	International Journal of Data Science and Analysis (Volume 6, Issue 4)
DOI	10.11648/j.ijdsa.20200604.13
Page(s)	113-119
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2020. Published by Science Publishing Group

Keywords

Sequential Bayesian Analysis, Bernoulli Opinion Polls, Election Forecasting, Simulation-Approach

References

[1]	L. Bursztyn, D. Cantoni, P. Funk, and N. Yuchtman, “Polls, the Press, and Political Participation: The Effects of Anticipated Election Closeness on Voter Turnout,” National Bureau of Economic Research, 2017, doi: 10.3386/w23490.
[2]	W. F. Christensen and L. W. Florence, “Predicting presidential and other multistage election outcomes using state-level pre-election polls,” American Statistician, 2008, doi: 10.1198/000313008X267820.
[3]	H. Keun Lee and Y. Woon Kim, “Public opinion by a poll process: Model study and Bayesian view,” Journal of Statistical Mechanics: Theory and Experiment, 2018, doi: 10.1088/1742-5468/aabbc5.
[4]	L. F. Stoetzer, M. Neunhoeffer, T. Gschwend, S. Munzert, and S. Sternberg, “Forecasting Elections in Multiparty Systems: A Bayesian Approach Combining Polls and Fundamentals,” Political Analysis, 2019, doi: 10.1017/pan.2018.49.
[5]	D. P. CHRISTENSON and C. D. SMIDT, “Polls and Elections: Still Part of the Conversation: Iowa and New Hampshire’s Say within the Invisible Primary,” Presidential Studies Quarterly, 2012, doi: 10.1111/j.1741-5705.2012.03994.x.
[6]	P. Selb and S. Munzert, “Forecasting the 2013 german bundestag election using many polls and historical election results,” German Politics, 2016, doi: 10.1080/09644008.2015.1121454.
[7]	P. Selb and S. Munzert, “Forecasting the 2013 Bundestag Election Using Data from Various Polls,” SSRN Electronic Journal, 2013, doi: 10.2139/ssrn.2313845.
[8]	R. McDonald and X. Mao, “Forecasting the 2015 General Election with Internet Big Data: An Application of the TRUST Framework,” Working Papers, 2015.
[9]	C. Spike and P. Vernon, US election analysis 2016 : Media, voters and the campaign. 2016.
[10]	P. C. Ordeshook and T. R. Palfrey, “Agendas, Strategic Voting, and Signaling with Incomplete Information,” American Journal of Political Science, 1988, doi: 10.2307/2111131.
[11]	M. Haspel and H. Gibbs Knotts, “Location, location, location: Precinct placement and the costs of voting,” Journal of Politics, 2005, doi: 10.1111/j.1468-2508.2005.00329.x.
[12]	M. Henn and N. Foard, “Social differentiation in young people’s political participation: The impact of social and educational factors on youth political engagement in Britain,” Journal of Youth Studies, 2014, doi: 10.1080/13676261.2013.830704.
[13]	M. Henn, M. Weinstein, and S. Forrest, “Uninterested youth? Young people’s attitudes towards party politics in Britain,” Political Studies, 2005, doi: 10.1111/j.1467-9248.2005.00544.x.
[14]	A. M. Williams, C. Jephcote, H. Janta, and G. Li, “The migration intentions of young adults in Europe: A comparative, multilevel analysis,” 2018, doi: 10.1002/psp.2123.
[15]	J. D. Byers, “Is political popularity a random walk?” Applied Economics, 1991, doi: 10.1080/00036849100000045.
[16]	D. Byers, J. Davidson, and D. Peel, “Modelling political popularity: An analysis of long-range dependence in opinion poll series,” Journal of the Royal Statistical Society. Series A: Statistics in Society, 1997, doi: 10.1111/j.1467-985X.1997.00075.x.
[17]	J. B. Carlin, “A case study on the choice, interpretation and checking of multilevel models for longitudinal binary outcomes,” Biostatistics, 2001, doi: 10.1093/biostatistics/2.4.397.
[18]	A. Gelman, Bayesian data analysis Gelman. 2013.
[19]	X. Puig and J. Ginebra, “A Bayesian cluster analysis of election results,” Journal of Applied Statistics, 2014, doi: 10.1080/02664763.2013.830088.
[20]	K. Lock and A. Gelman, “Bayesian combination of state polls and election forecasts,” Political Analysis, 2010, doi: 10.1093/pan/mpq002.
[21]	J. Mwanyekange, S. M. Mwalili, and O. Ngesa, “Bayesian Joint Models for Longitudinal and Multi-state Survival Data,” International Journal of Statistics and Probability, 2019, doi: 10.5539/ijsp.v8n2p34.
[22]	P. Selb and S. Munzert, “Estimating constituency preferences from sparse survey data using auxiliary geographic information,” Political Analysis, 2011, doi: 10.1093/pan/mpr034.
[23]	J. O. Berger, “Robust Bayesian analysis: sensitivity to the prior,” Journal of Statistical Planning and Inference, 1990, doi: 10.1016/0378-3758(90)90079-A.
[24]	M. Ghosh and J. Berger, “Stastical Decision Theory and Bayesian Analysis.,” Journal of the American Statistical Association, 1988, doi: 10.2307/2288950.
[25]	J. O. Berger, “Bayesian Analysis: A Look at Today and Thoughts of Tomorrow,” Journal of the American Statistical Association, 2000, doi: 10.1080/01621459.2000.10474328.
[26]	J. O. Berger, “Bayesian analysis: A look at today and thoughts of tomorrow,” in Statistics in the 21st Century, 2001.
[27]	D. Barber, Bayesian Reasoning and Machine Learning. 2011.

Cite This Article

Plain Text BibTeX RIS

APA Style

Jeremiah Kiingati, Samuel Mwalili, Anthony Waititu. (2020). Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach. International Journal of Data Science and Analysis, 6(4), 113-119. https://doi.org/10.11648/j.ijdsa.20200604.13

Copy | Download

ACS Style

Jeremiah Kiingati; Samuel Mwalili; Anthony Waititu. Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach. Int. J. Data Sci. Anal. 2020, 6(4), 113-119. doi: 10.11648/j.ijdsa.20200604.13

Copy | Download

AMA Style

Jeremiah Kiingati, Samuel Mwalili, Anthony Waititu. Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach. Int J Data Sci Anal. 2020;6(4):113-119. doi: 10.11648/j.ijdsa.20200604.13

Copy | Download

@article{10.11648/j.ijdsa.20200604.13,
  author = {Jeremiah Kiingati and Samuel Mwalili and Anthony Waititu},
  title = {Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach},
  journal = {International Journal of Data Science and Analysis},
  volume = {6},
  number = {4},
  pages = {113-119},
  doi = {10.11648/j.ijdsa.20200604.13},
  url = {https://doi.org/10.11648/j.ijdsa.20200604.13},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijdsa.20200604.13},
  abstract = {In this paper we apply sequential Bayesian approach to compare the outcome of the presidential polls in Kenya. We use the previous polls to form the prior for the current polls. Even though several authors have used non-Bayesian models for countrywide polling data to forecast the outcome of the presidential race we propose a Bayesian approach in this case. As such the question of how to treat the previous and current pre-election polls data is inevitable. Some researchers consider only the most recent poll others Combine all previous polls up the present time and treat it as a single sample, weighting only by sample size, while others Combine all previous polls but adjust the sample size according to a weight function depending on the day the poll is taken. In this paper we apply a sequential Bayesian model (as an advancement of the latter which is time sensitive) where the previous measure is used as the prior of the current measure. Our concern is to model the proportion of votes between two candidates, incumbent and challenger. A Bayesian model of our binomial variable of interest will be applied sequentially to the Kenya opinion poll data sets in order to arrive at a posterior probability statement. The simulation results show that the eventual winner must lead consistently and constantly in at least 60% of the opinions polls. In addition, a candidate demonstrating high variability is more likely to lose the polls.},
 year = {2020}
}

Copy | Download

TY  - JOUR
T1  - Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach
AU  - Jeremiah Kiingati
AU  - Samuel Mwalili
AU  - Anthony Waititu
Y1  - 2020/09/19
PY  - 2020
N1  - https://doi.org/10.11648/j.ijdsa.20200604.13
DO  - 10.11648/j.ijdsa.20200604.13
T2  - International Journal of Data Science and Analysis
JF  - International Journal of Data Science and Analysis
JO  - International Journal of Data Science and Analysis
SP  - 113
EP  - 119
PB  - Science Publishing Group
SN  - 2575-1891
UR  - https://doi.org/10.11648/j.ijdsa.20200604.13
AB  - In this paper we apply sequential Bayesian approach to compare the outcome of the presidential polls in Kenya. We use the previous polls to form the prior for the current polls. Even though several authors have used non-Bayesian models for countrywide polling data to forecast the outcome of the presidential race we propose a Bayesian approach in this case. As such the question of how to treat the previous and current pre-election polls data is inevitable. Some researchers consider only the most recent poll others Combine all previous polls up the present time and treat it as a single sample, weighting only by sample size, while others Combine all previous polls but adjust the sample size according to a weight function depending on the day the poll is taken. In this paper we apply a sequential Bayesian model (as an advancement of the latter which is time sensitive) where the previous measure is used as the prior of the current measure. Our concern is to model the proportion of votes between two candidates, incumbent and challenger. A Bayesian model of our binomial variable of interest will be applied sequentially to the Kenya opinion poll data sets in order to arrive at a posterior probability statement. The simulation results show that the eventual winner must lead consistently and constantly in at least 60% of the opinions polls. In addition, a candidate demonstrating high variability is more likely to lose the polls.
VL  - 6
IS  - 4
ER  -

Copy | Download

Author Information

Jeremiah Kiingati

Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya
Samuel Mwalili

Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya
Anthony Waititu

Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya

Download PDF

Sections

Plain Text BibTeX RIS

APA Style

Jeremiah Kiingati, Samuel Mwalili, Anthony Waititu. (2020). Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach. International Journal of Data Science and Analysis, 6(4), 113-119. https://doi.org/10.11648/j.ijdsa.20200604.13

Copy | Download

ACS Style

Jeremiah Kiingati; Samuel Mwalili; Anthony Waititu. Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach. Int. J. Data Sci. Anal. 2020, 6(4), 113-119. doi: 10.11648/j.ijdsa.20200604.13

Copy | Download

AMA Style

Jeremiah Kiingati, Samuel Mwalili, Anthony Waititu. Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach. Int J Data Sci Anal. 2020;6(4):113-119. doi: 10.11648/j.ijdsa.20200604.13

Copy | Download

@article{10.11648/j.ijdsa.20200604.13,
  author = {Jeremiah Kiingati and Samuel Mwalili and Anthony Waititu},
  title = {Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach},
  journal = {International Journal of Data Science and Analysis},
  volume = {6},
  number = {4},
  pages = {113-119},
  doi = {10.11648/j.ijdsa.20200604.13},
  url = {https://doi.org/10.11648/j.ijdsa.20200604.13},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijdsa.20200604.13},
  abstract = {In this paper we apply sequential Bayesian approach to compare the outcome of the presidential polls in Kenya. We use the previous polls to form the prior for the current polls. Even though several authors have used non-Bayesian models for countrywide polling data to forecast the outcome of the presidential race we propose a Bayesian approach in this case. As such the question of how to treat the previous and current pre-election polls data is inevitable. Some researchers consider only the most recent poll others Combine all previous polls up the present time and treat it as a single sample, weighting only by sample size, while others Combine all previous polls but adjust the sample size according to a weight function depending on the day the poll is taken. In this paper we apply a sequential Bayesian model (as an advancement of the latter which is time sensitive) where the previous measure is used as the prior of the current measure. Our concern is to model the proportion of votes between two candidates, incumbent and challenger. A Bayesian model of our binomial variable of interest will be applied sequentially to the Kenya opinion poll data sets in order to arrive at a posterior probability statement. The simulation results show that the eventual winner must lead consistently and constantly in at least 60% of the opinions polls. In addition, a candidate demonstrating high variability is more likely to lose the polls.},
 year = {2020}
}

Copy | Download

TY  - JOUR
T1  - Sequential Bayesian Analysis of Bernoulli Opinion Polls; a Simulation-Based Approach
AU  - Jeremiah Kiingati
AU  - Samuel Mwalili
AU  - Anthony Waititu
Y1  - 2020/09/19
PY  - 2020
N1  - https://doi.org/10.11648/j.ijdsa.20200604.13
DO  - 10.11648/j.ijdsa.20200604.13
T2  - International Journal of Data Science and Analysis
JF  - International Journal of Data Science and Analysis
JO  - International Journal of Data Science and Analysis
SP  - 113
EP  - 119
PB  - Science Publishing Group
SN  - 2575-1891
UR  - https://doi.org/10.11648/j.ijdsa.20200604.13
AB  - In this paper we apply sequential Bayesian approach to compare the outcome of the presidential polls in Kenya. We use the previous polls to form the prior for the current polls. Even though several authors have used non-Bayesian models for countrywide polling data to forecast the outcome of the presidential race we propose a Bayesian approach in this case. As such the question of how to treat the previous and current pre-election polls data is inevitable. Some researchers consider only the most recent poll others Combine all previous polls up the present time and treat it as a single sample, weighting only by sample size, while others Combine all previous polls but adjust the sample size according to a weight function depending on the day the poll is taken. In this paper we apply a sequential Bayesian model (as an advancement of the latter which is time sensitive) where the previous measure is used as the prior of the current measure. Our concern is to model the proportion of votes between two candidates, incumbent and challenger. A Bayesian model of our binomial variable of interest will be applied sequentially to the Kenya opinion poll data sets in order to arrive at a posterior probability statement. The simulation results show that the eventual winner must lead consistently and constantly in at least 60% of the opinions polls. In addition, a candidate demonstrating high variability is more likely to lose the polls.
VL  - 6
IS  - 4
ER  -

Copy | Download