Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion
In most practical applications, the quality of count data is often compromised due to errors-in-variables (EIVs). In this paper, we apply Bayesian approach to reduce bias in estimating the parameters of count data regression models that have mismeasured independent variables. Furthermore, the exposu...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Taylor & Francis
2018
|
| Subjects: |
| _version_ | 1825721579997233152 |
|---|---|
| author | Rozliman, Nur Aainaa Ibrahim, Adriana Irawati Nur Yunus, Rossita Muhamad |
| author_facet | Rozliman, Nur Aainaa Ibrahim, Adriana Irawati Nur Yunus, Rossita Muhamad |
| author_sort | Rozliman, Nur Aainaa |
| collection | UM |
| description | In most practical applications, the quality of count data is often compromised due to errors-in-variables (EIVs). In this paper, we apply Bayesian approach to reduce bias in estimating the parameters of count data regression models that have mismeasured independent variables. Furthermore, the exposure model is misspecified with a flexible distribution, hence our approach remains robust against any departures from normality in its true underlying exposure distribution. The proposed method is also useful in realistic situations as the variance of EIVs is estimated instead of assumed as known, in contrast with other methods of correcting bias especially in count data EIVs regression models. We conduct simulation studies on synthetic data sets using Markov chain Monte Carlo simulation techniques to investigate the performance of our approach. Our findings show that the flexible Bayesian approach is able to estimate the values of the true regression parameters consistently and accurately. |
| first_indexed | 2024-03-06T05:52:30Z |
| format | Article |
| id | um.eprints-20861 |
| institution | Universiti Malaya |
| last_indexed | 2024-03-06T05:52:30Z |
| publishDate | 2018 |
| publisher | Taylor & Francis |
| record_format | dspace |
| spelling | um.eprints-208612019-04-09T07:58:40Z http://eprints.um.edu.my/20861/ Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion Rozliman, Nur Aainaa Ibrahim, Adriana Irawati Nur Yunus, Rossita Muhamad Q Science (General) QA Mathematics In most practical applications, the quality of count data is often compromised due to errors-in-variables (EIVs). In this paper, we apply Bayesian approach to reduce bias in estimating the parameters of count data regression models that have mismeasured independent variables. Furthermore, the exposure model is misspecified with a flexible distribution, hence our approach remains robust against any departures from normality in its true underlying exposure distribution. The proposed method is also useful in realistic situations as the variance of EIVs is estimated instead of assumed as known, in contrast with other methods of correcting bias especially in count data EIVs regression models. We conduct simulation studies on synthetic data sets using Markov chain Monte Carlo simulation techniques to investigate the performance of our approach. Our findings show that the flexible Bayesian approach is able to estimate the values of the true regression parameters consistently and accurately. Taylor & Francis 2018 Article PeerReviewed Rozliman, Nur Aainaa and Ibrahim, Adriana Irawati Nur and Yunus, Rossita Muhamad (2018) Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion. Journal of Statistical Computation and Simulation, 88 (2). pp. 203-220. ISSN 0094-9655, DOI https://doi.org/10.1080/00949655.2017.1381845 <https://doi.org/10.1080/00949655.2017.1381845>. https://doi.org/10.1080/00949655.2017.1381845 doi:10.1080/00949655.2017.1381845 |
| spellingShingle | Q Science (General) QA Mathematics Rozliman, Nur Aainaa Ibrahim, Adriana Irawati Nur Yunus, Rossita Muhamad Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion |
| title | Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion |
| title_full | Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion |
| title_fullStr | Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion |
| title_full_unstemmed | Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion |
| title_short | Bayesian approach to errors-in-variables in count data regression models with departures from normality and overdispersion |
| title_sort | bayesian approach to errors in variables in count data regression models with departures from normality and overdispersion |
| topic | Q Science (General) QA Mathematics |
| work_keys_str_mv | AT rozlimannuraainaa bayesianapproachtoerrorsinvariablesincountdataregressionmodelswithdeparturesfromnormalityandoverdispersion AT ibrahimadrianairawatinur bayesianapproachtoerrorsinvariablesincountdataregressionmodelswithdeparturesfromnormalityandoverdispersion AT yunusrossitamuhamad bayesianapproachtoerrorsinvariablesincountdataregressionmodelswithdeparturesfromnormalityandoverdispersion |