Showing 1–3 of 3 results for author: Fuchs, C

stochprofML: Stochastic Profiling Using Maximum Likelihood Estimation in R

Authors: Lisa Amrhein, Christiane Fuchs

Abstract: Tissues are often heterogeneous in their single-cell molecular expression, and this can govern the regulation of cell fate. For the understanding of development and disease, it is important to quantify heterogeneity in a given tissue. We introduce the \proglang{R} package \pkg{stochprofML} which is designed to parameterize heterogeneity from the cumulative expression of small random pools of cells… ▽ More Tissues are often heterogeneous in their single-cell molecular expression, and this can govern the regulation of cell fate. For the understanding of development and disease, it is important to quantify heterogeneity in a given tissue. We introduce the \proglang{R} package \pkg{stochprofML} which is designed to parameterize heterogeneity from the cumulative expression of small random pools of cells. This method outweighs the demixing of mixed samples with a saving in cost and effort and less measurement error. The approach uses the maximum likelihood principle and was originally presented in Bajikar et al.(2014); its extension to varying pool sizes was used in Tirier et al. (2019). We evaluate the algorithm's performance in simulation studies and present further application opportunities. △ Less

Submitted 19 April, 2020; originally announced April 2020.

Comments: 52 pages, 24 figures

Bayesian Inference for Diffusion Processes: Using Higher-Order Approximations for Transition Densities

Authors: Susanne Pieschner, Christiane Fuchs

Abstract: Modelling random dynamical systems in continuous time, diffusion processes are a powerful tool in many areas of science. Model parameters can be estimated from time-discretely observed processes using Markov chain Monte Carlo (MCMC) methods that introduce auxiliary data. These methods typically approximate the transition densities of the process numerically, both for calculating the posterior dens… ▽ More Modelling random dynamical systems in continuous time, diffusion processes are a powerful tool in many areas of science. Model parameters can be estimated from time-discretely observed processes using Markov chain Monte Carlo (MCMC) methods that introduce auxiliary data. These methods typically approximate the transition densities of the process numerically, both for calculating the posterior densities and proposing auxiliary data. Here, the Euler-Maruyama scheme is the standard approximation technique. However, the MCMC method is computationally expensive. Using higher-order approximations may accelerate it, but the specific implementation and benefit remain unclear. Hence, we investigate the utilisation and usefulness of higher-order approximations in the example of the Milstein scheme. Our study demonstrates that the MCMC methods based on the Milstein approximation yield good estimation results. However, they are computationally more expensive and can be applied to multidimensional processes only with impractical restrictions. Moreover, the combination of the Milstein approximation and the well-known modified bridge proposal introduces additional numerical challenges. △ Less

Submitted 15 September, 2020; v1 submitted 6 June, 2018; originally announced June 2018.

Comments: 33 pages, 15 figures

Discussion of: Statistical analysis of an archeological find

Authors: Camil Fuchs

Abstract: Discussion of ``Statistical analysis of an archeological find'' by Andrey Feuerverger [arXiv:0804.0079] Discussion of ``Statistical analysis of an archeological find'' by Andrey Feuerverger [arXiv:0804.0079] △ Less

Submitted 1 April, 2008; originally announced April 2008.

Comments: Published in at http://dx.doi.org/10.1214/08-AOAS99D the Annals of Applied Statistics (http://www.imstat.org/aoas/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Report number: IMS-AOAS-AOAS99D

Journal ref: Annals of Applied Statistics 2008, Vol. 2, No. 1, 57-65