dc.contributor | Cepeda Cuervo, Edilberto | |
dc.contributor | Inferencia Bayesiana | |
dc.creator | Zárate Solano, Héctor Manuel | |
dc.date.accessioned | 2022-02-05T00:31:26Z | |
dc.date.available | 2022-02-05T00:31:26Z | |
dc.date.created | 2022-02-05T00:31:26Z | |
dc.date.issued | 2022-01 | |
dc.identifier | https://repositorio.unal.edu.co/handle/unal/80887 | |
dc.identifier | Universidad Nacional de Colombia | |
dc.identifier | Repositorio Institucional Universidad Nacional de Colombia | |
dc.identifier | https://repositorio.unal.edu.co/ | |
dc.description.abstract | Statistical applications need to address an increasing complexity due to new data arising from recent technologies, new phenomenons, and diverse sources of uncertainty. The demand for flexible methods with non-standard data structures, high-dimensional real-time estimation, and latent models framework have caused semiparametric modeling to play a crucial role in contemporary statistical analysis. We provide flexible Bayesian methods to jointly infer the mean, variance, and skewness functions when the response variable comes either from a two-parameter exponential family or asymmetric distributions. Hence, we implemented Bayesian algorithms based on MCMC sampling techniques and deterministic variational Bayesian learning theory. In these settings, each sub-model depends on some covariates parametrically and for others in a non-parametrically way. It follows that understanding how the moments change with predictors is a goal of Statistics, and it is of intrinsic interest given the role in approximating other quantities. We propose several modeling scenarios that benefit from the fusion of the graphical models' approach to Bayesian semiparametric regression under the architecture of GLM models. The significance and implications of our strategy lie in its potential to contribute to a unified computational methodology that provides insight into many complex models that otherwise could be intractable analytically. Therefore, combining data models and algorithms contribute to solving real-world problems enjoying crucial advantages related to faster computation time, which allow not only to explore quickly many models for the data but to estimate them accurately. | |
dc.description.abstract | Las aplicaciones estadísticas deben abordar una complejidad cada vez mayor debido a los nuevos datos que surgen con las tecnologías recientes, los nuevos fenómenos y las diversas fuentes de incertidumbre. La demanda por métodos con estructuras de datos no estándar, estimación en tiempo real de alta dimensión y modelos latentes adecuados ha causado que los modelos semiparamétricos desempeñen un papel crucial en el análisis estadístico reciente. En esta tesis se implementan métodos Bayesianos flexibles para inferir conjuntamente las funciones de media, varianza y asimetría cuando la variable de respuesta proviene de la familia exponencial biparamétrica o de distribuciones asimétricas. La aproximación es obtenida con métodos basados en técnicas de simulación de Monte Carlo con cadenas de markov y en algoritmos de aprendizaje variacional determinístico. En estos escenarios, cada submodelo incluye variables en forma paramétrica y no paramétrica para analizar el efecto de los predictores sobre los momentos. Los escenarios de modelamiento se benefician de la fusión entre los modelos gráficos y la regresión semiparamétrica Bayesiana utilizando la arquitectura de modelos lineales generalizados. La importancia e implicaciones de nuestra estrategia radican en su potencial para contribuir con una metodología computacional unificada que proporciona información sobre una gran variedad de modelos complejos que, de otro modo, podrían resultar analíticamente intratables. Por lo tanto, la combinación de modelos de datos y algoritmos contribuye a resolver problemas del mundo real y disfruta de ventajas cruciales relacionadas con el bajo tiempo de cómputo, lo cual permite no solo explorar rápidamente muchos modelos para los datos, sino también estimarlos con precisión. (Texto tomado de la fuente). | |
dc.language | eng | |
dc.publisher | Universidad Nacional de Colombia | |
dc.publisher | Bogotá - Ciencias - Doctorado en Ciencias - Estadística | |
dc.publisher | Departamento de Estadística | |
dc.publisher | Facultad de Ciencias | |
dc.publisher | Bogotá, Colombia | |
dc.publisher | Universidad Nacional de Colombia - Sede Bogotá | |
dc.relation | Anderson, D. F. and Livingston, P. S. The zero-divisor graph of a commutative ring. Journal
of Algebra, 217(2):434–447, 1999. | |
dc.relation | Berry, S., Carroll, R., and Ruppert, D. Bayesian smoothing and regression splines for
measurement error problems. Journal of the American Statistical Association, 97(457):160
– 169, 2002. | |
dc.relation | Bishop, C. M. Pattern recognition and Machine Learning, volume 1 of Graduate Texts in
Mathematics. New York : Springer, 2006. | |
dc.relation | Blei, D. M., Kucukelbir, A., and McAuliffe, J.D. Variational inference: a review for
statisticians. Journal of the American Statistical Association, 112:859–857, 2017. | |
dc.relation | Brooks, S., Gelman, A., Jones, G., and Meng, X. Handbook of Markov Chain Monte Carlo.
Handbooks of Modern and Statistical Methods. Chapman Hall/CR, 2011. | |
dc.relation | Cepeda, E. and Gamerman, D. Bayesian modeling of variance heterogeneity in normal
regression models. J. Prob. Stat, 14:207–221, 2001. | |
dc.relation | Crainiceanu, C., Ruppert, D., and Wand, M. Bayesian analysis for penalized spline regression
using winbugs. Journal of Statistical Software, 14:1–24, 2005. | |
dc.relation | Crevar, M. Shared file systems: Determining the best choice for your distributed SAS®
foundation applications. SAS Institute Inc., Cary, NC., pages 1–11, 2017.
21 | |
dc.relation | Dey, D.K., Gelfand, A.E., and Peng, F. Overdispersed generalized linear models. Journal of
Statistical Planning and Inference,, 64(64):93–108, 1997. | |
dc.relation | Evans, M. J. and Rosenthal, J. S. Probability and Statistics: The science of Uncertainty. The
American Statistician. New York: W.H Freeman and Company, 2004. | |
dc.relation | Gilks, W.R., Richardson, S., and Spiegelhalter, D.J. Markov Chain Monte Carlo in practice.
Interdisciplinary Statistics. Chapman Hall/CR, 1998. | |
dc.relation | Green, P.J. and Silverman, B.W. Nonparametric Regression and Generalized Linear Models:
A roughness penalty approach. Chapman and Hall, London, 1994. | |
dc.relation | Groll, A., Hambuckers, J., , Kneib, T., and Umlauf, N. Lasso-type penalization in the
framework of generalized additive models for location, scale and shape. Computational
Statistics Data Analysis, 140:59–73, 2019. | |
dc.relation | Hastings, W. Monte carlo sampling methods using markov chains and their applications.
Biometrika, (57):97–109, 1970. | |
dc.relation | Jang, E. A beginner’s guide to variational methods: Mean-field approximation.
http://blog.evjang.com/2016/08/variational-bayes.html, (3):233–240, 2016. | |
dc.relation | Kullback, S. Information Theory and Statistics. Gloucester , MASS, 1978. | |
dc.relation | Landau, D. and Binder, K. A Guide to Monte Carlo Simulations in Statistical Physics.
Cambridge University Press, 2005. | |
dc.relation | Nakajima, S., Watanabe, K., and Sugiyama, M. Variational Bayesian Learning Theory,
volume 1 of Graduate Texts in Mathematics. Cambridge University Press, 2019. | |
dc.relation | Nosedal-Sanchez, A., Storlie, C., Thomas, C., and Chisrensen, R. Reproducing kernel hilbert
spaces for penalized regression : A tutorial. The American Statistician, (66):50–60, 2012. | |
dc.relation | Ormerod, J. T. and Wand, J. T. Explaining variational approximations. The American
Statistician, 64:140–153, 2001.
22 | |
dc.relation | Pierce, N. and Wand, D. Penalized splines and reproducible kernel methods. American
Statistical Association, (3):233–240, 2006. | |
dc.relation | Ruppert, D., Wand, M., and Carroll, R. Semiparametric regression during 2003-2007.
Electronic Journal of Statistics, 3:1193–1256, 2009. | |
dc.relation | Stasinopoulos, D. and Rigby, R. Generalized additive models for location scale and shape
(gamlss) in r. Journal of Statistical Software, 23:1–46, 2007. | |
dc.relation | Turkman, M., Paulino, C., and Muller, P. Computational Bayesian Statistics . An Introduction.
Cambridge, 2019. | |
dc.relation | Umlauf, N., Nadja, K., and Achim, Z. Bamlss: Bayesian additive models for location, scale,
and shape (and beyond). Journal of Computational and Graphical Statistics, 27:612–627, 2018. | |
dc.relation | Wahba, G. Spline Models for Observational Data. Society for Industrial and Applied
Mathematics, 1990. | |
dc.relation | Whaba, G. and Wendelberger, J. Some new mathematical methods for variational objective
analysis using splines and cross validation. Monthly weather review, 108:1122–1143, 1980.
23 | |
dc.relation | Berry, S., Carroll, R., and Ruppert, D. Bayesian smoothing and regression splines for
measurement error problems. Journal of the American Statistical Association, (457):160–169,
2011. | |
dc.relation | Cepeda, E. Variability modeling in Generalized Linear models. PhD thesis, Unpublished Ph.D
thesis, Mathematics Institute Universidade Federal do Rio de Janeiro, 2001. | |
dc.relation | Cepeda, E. and Gamerman, D. Bayesian methodology for modeling parameters in the two
parametric exponential family. Estadística, 57:93–105, 2005. | |
dc.relation | Cepeda,E., Achcar,J., and Garrido Lopera,L. Bivariate beta regression models: joint modeling
of the mean, dispersion and association parameters. Journal of Applied statistics, 41(3):677–
687, Marzo 2014. | |
dc.relation | Crainiceanu, C. Spatially adaptative bayesian penalized splines with heteroscedastic errors.
Journal of Computational and Graphical Statistics, (2):265–288, 2007. | |
dc.relation | Currie, I. and Burban, M. Flexible smoothing with p-splines : a unified approach. Statistical
Modelling, (4):333–349, 2002. | |
dc.relation | Davidian, M., Lin, X., Morris, J., and Stefanski, O. The Work of Raymond J. Carroll. The
impact and influence of a Statistician. 2014.
49 | |
dc.relation | Denison, D., Mallick, B., and Smith, F. A bayesian cart algorithm. Biometrika, (2):363 – 367,
1998. | |
dc.relation | Dey, D.K., Gelfand,A.E., and Peng, F. Overdispersed generalized linear models. Journal of
Statistical Planning and Inference,, 64(64):93–108, 1997. | |
dc.relation | Eilers, P.and Marx, B. and Durbán, M. Twenty years of p-splines. SORT (Statistics and
Operations Research Transactions), (39), 2014. | |
dc.relation | Gamerman, D. Sampling from the posterior distribution in generalized linear mixed models.
Instituto de matemática, Universidade Federal do Rio de Janeiro, pages 59 – 68, 1997. | |
dc.relation | Gijbels, I. and Prosdocimi, I. Flexible mean and dispersion function estimation in extended
generalized additive models. Communications in statistics - Theory and Methods, (41):3259 –
3277, 2012. | |
dc.relation | Gu,C. Smoothing Spline ANOVA Models. Springer, West Lafayette,USA, 2002. | |
dc.relation | Littell, R. and Schabenberger, O. SAS for Mixed Models. Number 2. 2006. | |
dc.relation | Loomis, C. MCMC in SAS: From scratch or by proc. Western users of SAS software 2016, 1(1):1
– 19, 2016. | |
dc.relation | Ma, Y. and Carroll R.J. Locally efficient estimators for semiparametric models with
measurement error. Journal of the American Statistical Association, (101):1465–1474, 2006. | |
dc.relation | Mencitas, M. and Wand, M. Variational inference for heteroscedastic semiparametric
regression. School of mathematical sciences, University of Technology. Sydney, Australia, 2014. | |
dc.relation | Nosedal-Sanchez, A., Storlie, C., Thomas, C., and Chisrensen, R. Reproducing kernel hilbert
spaces for penalized regression : A tutorial. The American Statistician, (66):50–60, 2012. | |
dc.relation | Nott, D. Semiparametric estimation of mean and variance functions for non-gaussian data.
Computational Statistics, (3-4):603–620, 2006. | |
dc.relation | Pierce, N. and Wand, D. Penalized splines and reproducible kernel methods. American
Statistical Association, (3):233–240, 2006. | |
dc.relation | Pinheiro, J. and Bates, D. Mixed-Effects Models in S and S-Plus. Springer Verlag, 2009. | |
dc.relation | Robert, C.P., Elvira,V., Tawn, N., and Wu, C. Accelerating mcmc algorithms. Journal of the
American Statistical Association, 2018. | |
dc.relation | Ruppert, D., Wand, M.P., Holst,U., and Hössjer, O. Local polynomial variance-function
estimation. Technometrics, (39):262–273, 1997. | |
dc.relation | Spiegelhalter, D. and Best, N. Bayesian approaches to multiple sources of evidence in complex
cost-effectiveness modelling. Statistics in Medicine, (23):3687 – 3709, 2003. | |
dc.relation | Tran, M., Nguyen, N., Nott, D., and Kohn, R. Bayesian deep net glm and glmm. SORT (
arXiv:1805.10157v1 [stat.CO]), 2018. | |
dc.relation | Xu,D. and Zhang,Z. A semiparametric bayesian approach to joint mean and variance models.
Statistics & Probability Letters, 83(7):1624 – 1631, 2013. | |
dc.relation | Azzalini, A. A class of distribution which includes the normal ones. Scandinavian Journal of
Statistics, 12:171–178, 1985. | |
dc.relation | Azzalini, A. Further results on a class of distributions which includes the normal ones.
Statistica, 46:199–208, 1986. | |
dc.relation | Charenza, W. and Diaz, C. Choosing the rigth skew normal distribution: the macroeconomist
dilemma. Journal of Forecasting, 19:235–254, 2015. | |
dc.relation | Dursun, A. and Ersin, Y. Modified estimators in semiparametric regression models with rightcensored
data. Journal of Statistical Computation and Simulation, 88:1470–1498, 2018. | |
dc.relation | Ferreira, J. and Steel, M. A new class of skewed multivariate distributions with applications
to regression analysis. Statistica Sinica, 17:505–529, 2007. | |
dc.relation | Franceschini, C. and Loperfido, N. Testing for normality when the sampled distribution is
extended skew-normal. Mathematical and statistical methods for actuarial sciences and finance
Springer, 2014. | |
dc.relation | Genton, M. Discussion of the ’skew-normal’. Scandinavian Journal of Statistics, 32:189–198,
2005. | |
dc.relation | Groll, A., Hambuckers, J., , Kneib, T., and Umlauf, N. Lasso-type penalization in the
framework of generalized additive models for location, scale and shape. Computational
Statistics Data Analysis, 140:59–73, 2019. | |
dc.relation | Gómez, H., Salinas, H., and Bolfarine, H. Generalized skew-normal models: properties and
inference. Statistics, 40:495–505, 2006. | |
dc.relation | Huiqiong, L., Liucang, W., and Ting, M. Variable selection in joint location, scale and skewness
models of the skew-normal distribution. J Syst Sci Complex, 30:694–709, 2017. | |
dc.relation | Kuligowsk, A. and Mendes, L. Working with sparse matrices in sas®. Proceedings of the
SAS® Global Forum 2019 Conference, pages 1–9, 2019. | |
dc.relation | Ma, Y. and Genton, M. Flexible class of skew-symmetric distributions. Scandinavian Journal
of Statistics, 31:459–468, 2004. | |
dc.relation | Ma, Y., Genton, M., and Tsiatis, A. Locally efficient semiparametric estimators for generalized
skew-elliptical distributions. Journal of American Statistical Association, 100:980–989, 2005. | |
dc.relation | Perez, P., Acosta, R., and Perez, S. A bayesian genomic regression model with skew normal
random errors? G3, 8:1771–1785, 2018. | |
dc.relation | Potgieter, C. and Genton, M. Bayesian analysis of two-piece normal regression models.
Presented at Joint Statistical meeting, San Francisco Statistics, 2003. | |
dc.relation | Potgieter, C. and Genton, M. Characteristic function-based semiparametric inference for skewsymmetric
models. Scandinavian Journal of Statistics, 40:1803–1819, 2013. | |
dc.relation | Satori, N. Bias prevention of maximum likelihood estimates for scalar skew normal and skew
t distributions. Journal of Statistical Planning and Inference, 136:4259–4275, 2006. | |
dc.relation | Stasinopoulos, D. and Rigby, R. Generalized additive models for location scale and shape
(gamlss) in r. Journal of Statistical Software, 23:1–46, 2007. | |
dc.relation | Umlauf, N., Nadja, K., and Achim, Z. Bamlss: Bayesian additive models for location, scale,
and shape (and beyond). Journal of Computational and Graphical Statistics, 27:612–627, 2018. | |
dc.relation | Wilhelmsson, Y. Value at risk with time varying variance, skewness and kurtosis - the nig-acd
model. Econometrics Journal, 12:82–104, 2009. | |
dc.relation | Yu, K., Alhamzawi, A., Becker, F., and Lord, J. Statistical methods for body mass index: a
selective review of the literature. arXiv:1412.3653v1 [stat.AP], pages 1–32, 2014. | |
dc.relation | Zareifard, H. and Khaledi, M. Non-gaussian modeling of spatial data using scale mixing of a
unified skew gaussian process. Journal of Multivariate Analysis, 114:16–28, 2013.
99 | |
dc.relation | Ahmedt, D., Ali, M., Denman, S., Fookes, C., and Petersson, L. Graph-based deep learning for
medical diagnosis and analysis: Past, present and future. arXIV:2105.13137v1 cs.LG, pages
1–41, 2021. | |
dc.relation | Barro, S and Sala i Martin, X. Economic growth. MIT, 2004. | |
dc.relation | Bishop, C. Pattern recognition and machine learning. Springer, 2016. | |
dc.relation | Bugbee, B., Bredit, J., and Van der Woerd, M. Laplace variational approximation for
semiparametric regression in presence of heterocedastic errors. Journal of Computational and
Graphical Statistics, 25:225–245, 2016. | |
dc.relation | Buxton, R. Introduction to Functional Magnetic Resonance Imaging : Principles and
Techniques. Cambridge, 2009. | |
dc.relation | Caffo, B., Bowman, D., Elberly, L., and Bassett, S. Handbook of Markov Chain Monte Carlo:
Part II, chapter 14. Chapman Hall / CRC, 2011. | |
dc.relation | Faes, C. and Wand, M.P. Semiparametric mean field variational bayes: General principles and
numerical issues. Journal of Machine Learning Research, (17):1–47, 2016. | |
dc.relation | Hans, S. MRI made easy (...well almost). Schering, 1990. | |
dc.relation | Koller, D. and Friedman, N. Probabilistic Graphical Models: Principles and Techniques. The
MIT Press, 2010. | |
dc.relation | Larsen, W. and McCleary, S. The use of partial residual plots in regression analysis.
Technometrics, (14):781–790, 1970. | |
dc.relation | Lazaro, M. Bayesian warped gaussian processes. Advances in Neural Information Processing
Systems, 26:225–245, 2013. | |
dc.relation | Mencitas, M. and Wand, M. Variational inference for heteroscedastic semiparametric
regression. School of mathematical sciences, University of Technology. Sydney, Australia, 2014. | |
dc.relation | Nakajima, S., Watanabe, K., and Sugiyama, M. Variational Bayesian Learning Theory.
Cambridge University press, 2019. | |
dc.relation | Nott , D., Tran, M., and Kuk, A. Efficient variational inference for generalized linear mixed
models with large datasets. arXiv preprint, 2018. | |
dc.relation | Potgieter, C. and Genton, M. Bayesian analysis of two-piece normal regression models.
Presented at Joint Statistical meeting, San Francisco Statistics, 2003. | |
dc.relation | Rindler, F. Calculus of Variations. Springer-Verlag, 2016. | |
dc.relation | Starke, L. and Ostwald, D. Variational bayesian parameter estimation techniques for the
general linear model. Frontiers in Neuroscience, pages 1–22, 2017. | |
dc.relation | Zárate, H. and Cepeda, E. Semiparametric smoothing spline in joint mean and dispersion
models with responses from the biparametric exponential family: A bayesian perspective.
Statistics, Optimization Information Computing, 9(2):351–367, 2021. | |
dc.rights | Reconocimiento 4.0 Internacional | |
dc.rights | http://creativecommons.org/licenses/by/4.0/ | |
dc.rights | info:eu-repo/semantics/openAccess | |
dc.title | Semiparametric smoothing spline to joint mean and variance models with responses from the biparametric exponential family: a bayesian perspective | |
dc.type | Trabajo de grado - Doctorado | |