Using a posterior distribution of Bayesian LASSO, we construct a semi-norm on the parameter space. We show that the partition function depends on the ratio of l¹ and l² norms. We derive the concentration of Bayesian LASSO, and present MCMC convergence diagnosis.

LASSO, Bayes, MCMC, log-concave, geometry, incomplete gamma function.

Received: March 29, 2022; Revised: April 21, 2022; Accepted: May 5, 2022; Published: May 30, 2022

How to cite this article: Daoud Ounaissi and Nadji Rahmania, Bayesian LASSO: concentration and MCMC diagnosis, Far East Journal of Theoretical Statistics 65 (2022), 35-54. http://dx.doi.org/10.17654/0972086322005

This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License

References
[1] R. Alhamza and H. T. Mohammad Ali, A new Gibbs sampler for Bayesian lasso, Comm. Statist. Simulation Comput. 49(7) (2020), 1855 1871.
[2] K. Ball, Logarithmically concave functions and sections of convex sets in Studia Math. 88 (1988), 69-84.
[3] A. Beck and M. Teboulle, A fast iterative shrinkage-thresholding algorithm for linear inverse problem, SIAM J. Imag. Sci. 2 (2009), 183-202.
[4] S. Chakrabort, S. K. Bhattacharya and K. Khare, Estimating accuracy of the MCMC variance estimator: asymptotic normality for batch means estimators, Statist. Probab. Lett. 183 (2022), Paper No. 109337, 6 pp.
[5] S. Chen, D. L. Donoho and M. Saunders, Atomic decomposition by basis pursuit, SIAM J. Sci. Comput. 20(1) (1998), 33-61.
[6] E. T. Copson, Asymptotic Expansions, Cambridge University Press, 1965.
[7] I. Daubechies, M. Defrise and C. De Mol, An iterative thresholding algorithm for linear inverse problems with a sparsity constraint, Comm. Pure Appl. Math. 57(11) (2004), 1413-1457.
[8] A. Dermoune, D. Ounaissi and N. Rahmania, Oscillation of Metropolis-Hastings and simulated annealing algorithms around LASSO estimator, Math. Comput. Simulation 135 (2017), 39-50.
[9] B. Efron, T. Hastie, I. Johnstone and R. Tibshirani, Least angle regression, Ann. Statist. 37 (2004), 407-499.
[10] A. Erdélyi, Higher transcendental functions, California Institute of Technology, Bateman Manuscript Project, Vol. 2, 1953, p. 119.
[11] G. Fort, S. Le Corff, E. Moulines and A. Schreck, A shrinkage-thresholding Metropolis adjusted Langevin algorithm for Bayesian variable selection, 2015. arXiv:1312.5658v3 [math.ST].
[12] J. Johndrow, P. Orenstein and A. Bhattacharya, Scalable approximate MCMC algorithms for the horseshoe prior, J. Mach. Learn. Res. 21 (2020), 1-61.
[13] B. Klartag and V. D. Milman, Geometry of log-concave functions and measures, Geom. Dedicata 112(1) (2005), 169-182.
[14] K. Mengersen and R. L. Tweedie, Rates of convergence of the Hastings and Metropolis algorithms, Ann. Statist. 24 (1994), 101-121.
[15] P. Natalini and B. Palumbo, Inequalities for the incomplete gamma function, Math. Inequal. Appl. 3(1) (2000), 69-77.
[16] N. Parikh and S. Boyed, Proximal algorithms, Found. Trends Optim. 1(3) (2003), 123-231.
[17] M. Pereyra, Proximal Markov chain Monte Carlo algorithms, Stat. Comput. 26 (2016), 745-760.
[18] G. O. Roberts and A. L. Tweedie, Geometric convergence and central limit theorems for multidimensional Hastings and Metropolis algorithms, Biometrika 83(1) (1996), 95-110.
[19] N. M. Temme, Numerical and asymptotic aspects of parabolic cylinder functions, J. Comput. Appl. Math. 121 (2000), 221-246.
[20] R. Tibshirani, Regression shrinkage and selection via Lasso, J. R. Stat. Soc. Ser. B Stat. Methodol. 58(1) (1996), 267-288.
[21] R. Tibshirani, The Lasso problem and uniqueness, Electron. J. Stat. 7 (2013), 1456-1490.