Dienstag, 22.10.2024, 12-13 Uhr in W9-109
apl. Prof. Dr. Odile Sauzet
Universität Bielefeld
Exploring the effects of dichotomisation in survival analysis and suggestion of a distributional approach
The limitations resulting from the dichtomisation of continuous outcome have been extensively described. But the need to present results based on binary outcomes in particular in health science remains. Alternatives based on the distribution of the continuous outcome have been proposed. Here we explore the possibilities of using a distributional approach in the context of time-to-event analysis when the event is the results of the dichotomisation of a continuous outcome. For this we propose in a first step a distributional version of the Kaplan-Meier estimate of the survival function.
Dienstag, 05.11.2024, 12-13 Uhr in W9-109
Jan-Ole Koslik
Universität Bielefeld
Efficient smoothness selection for nonparametric Markov-switching models via quasi restricted maximum likelihood estimation
Markov-switching models are powerful tools that allow capturing complex patterns from time series data driven by latent states. Recent work has highlighted the benefits of estimating components of these models nonparametrically, enhancing their flexibility and reducing biases, which in turn can improve state decoding, forecasting, and overall inference. Formulating such models using penalised splines is straightforward, but practically feasible methods for a data-driven smoothness selection in these models are still lacking. Traditional techniques, such as cross-validation and information criteria-based selection suffer from major drawbacks, most importantly their reliance on computationally expensive grid search methods, hampering practical usability for Markov-switching models. Michelot (2022) suggested treating spline coefficients as random effects with a multivariate normal distribution and using the R package TMB (Kristensen et al., 2015) for marginal likelihood maximisation. While this method avoids grid search and typically results in adequate smoothness selection, it entails a nested optimisation problem, thus being computationally demanding. We propose to exploit the simple structure of penalised splines treated as random effects, thereby greatly reducing the computational burden while potentially improving fixed effects parameter estimation accuracy. The proposed method offers a reliable and efficient mechanism for smoothness selection, rendering the estimation of Markov-switching models involving penalised splines feasible for complex data structures.
Dienstag, 19.11.2024, 12-13 Uhr in W9-109
Nayeli Gast Zepeda
Universität Bielefeld
Penalizing Infeasibility in Neural Combinatorial Optimization: An Experimental Study on Vehicle Routing Problems
Neural Combinatorial Optimization (NCO) methods have shown promise for vehicle routing problems (VRPs), with advances in problem scale and architectural improvements. However, these methods have primarily been tested on problems where feasible solutions are easily found. While previous work assumed neural networks could learn to respect constraints through penalty terms, we demonstrate experimentally that such penalty schemes fail to ensure solution feasibility, limiting the applicability of current NCO approaches to many real-world routing problems.
Dienstag, 03.12.2024, 12-13 Uhr in W9-109
Aktuelle Forschungsthemen im ZeSt
Dienstag, 17.12.2024, 12-13 Uhr in W9-109
Dr. Christoph Kiefer
Universität Bielefeld
Definition and Identification of Causal Ratio Effects
Dienstag, 14.01.2025, 12-13 Uhr in W9-109
Johannes Brachem
Georg-August-University Göttingen
Titel folgt
Dienstag, 28.01.2025, 12-13 Uhr in W9-109
Prof. Dr. Jan Gertheiss
Helmut-Schmidt-Universität
Titel folgt