Registered Data
Contents
- 1 [CT081]
- 1.1 [02011] Network models with truncated Poisson-Dirichlet process priors
- 1.2 [01132] Nonparametric Moment-based Estimation of Simulated Models via Regularized Regression
- 1.3 [00560] Semiparametric Kernel Estimation with Bayesian Bandwidths for Multivariate Nonnegative Data
- 1.4 [02249] An Adaptive Time Stepping Scheme for Rate-Independent Systems
- 1.5 [00887] Higher order Haar wavelet method for parabolic inverse problem
[CT081]
[02011] Network models with truncated Poisson-Dirichlet process priors
- Session Date & Time : 3E (Aug.23, 17:40-19:20)
- Type : Contributed Talk
- Abstract : We introduce a Bayesian nonparametric network model based on the truncated Poisson-Dirichlet process prior. In our model, the sociability parameters of the nodes are sorted in descending order. This enables us to focus on the most popular nodes of the network. We will show the simulation algorithm and posterior inference method for this model. Numerical implementations will also be discussed based on simulated observations and real-world datasets.
- Classification : 62G05, 62F15, 60G51
- Author(s) :
- Junyi Zhang (The Hong Kong Polytechnic University)
- Angelos Dassios (London School of Economics)
[01132] Nonparametric Moment-based Estimation of Simulated Models via Regularized Regression
- Session Date & Time : 3E (Aug.23, 17:40-19:20)
- Type : Contributed Talk
- Abstract : We exploits regularized regression to estimate the parameters of a simulation model. On each run of the model, we compute some simulated statistics, and we nonparametrically estimate the function linking the generated statistics and the associated parameters. We obtain estimates of the parameters through the previous estimate using the real-world statistics as explanatory variables. We characterize the asymptotic theory of the estimator and we evaluate the approach through a simulation study and an application.
- Classification : 62G08, 62E20, 62E17, 62P25
- Author(s) :
- Mario Martinoli (Sant'Anna School of Advanced Studies)
- Raffaello Seri (Università degli Studi dell'Insubria)
[00560] Semiparametric Kernel Estimation with Bayesian Bandwidths for Multivariate Nonnegative Data
- Session Date & Time : 3E (Aug.23, 17:40-19:20)
- Type : Contributed Talk
- Abstract : We introduce a flexible semiparametric kernel method for smoothing distributions on nonnegative supports. This multivariate estimator is guided by a given parametric part, here an uncorrelated exponential distribution estimated by maximum likelihood, and a nonparametric part which is a weight function to be smoothed through multiple gamma kernels. Also, a diagnostic model discusses the choice between the parametric, semiparametric and nonparametric approaches. Finally, practical multivariate semicontinuous datasets illustrate the usefulness of the method.
- Classification : 62Gxx, 62Hxx
- Author(s) :
- Sobom Matthieu Somé (Université Thomas SANKARA)
- Célestin C. Kokonendji (Université Bourgogne Franche-Comté)
[02249] An Adaptive Time Stepping Scheme for Rate-Independent Systems
- Session Date & Time : 3E (Aug.23, 17:40-19:20)
- Type : Contributed Talk
- Abstract : We investigate a local incremental stationary scheme for the numerical solution of rate-independent systems. The main novelty of our approach in comparison to existing methods is an adaptive choice of the step size for the update of the curve parameter. It is proven that the piecewise affine approximations generated by the algorithm converge (weakly) to a so-called parametrized balanced viscosity solution. Numerical experiments illustrate the theoretical findings and show a significant increase of the step size during sticking and in viscous jumps.
- Classification : 65J08, 65K15, 65M12, 65M50
- Author(s) :
- Merlin Andreia (TU Dortmund University)
- Christian Meyer (TU Dortmund University)
[00887] Higher order Haar wavelet method for parabolic inverse problem
- Session Date & Time : 3E (Aug.23, 17:40-19:20)
- Type : Contributed Talk
- Abstract : In the talk, we propose a higher order Haar wavelet method for solving parabolic inverse problem with a control parameter. Haar wavelets are used for the spatial discretization whereas backward Euler scheme is used for the temporal discretization. Stability results and error estimates are derived. The proposed method is tested on few examples and the results are compared with some existing literature. The proposed method is performing better in terms of absolute error and maximum error.
- Classification : 65T60, 65C30
- Author(s) :
- Gopal Priyadarshi (KIng Abdullah University of Science and Technology)