TY - JOUR A1 - Thapa, Samudrajit A1 - Park, Seongyu A1 - Kim, Yeongjin A1 - Jeon, Jae-Hyung A1 - Metzler, Ralf A1 - Lomholt, Michael A. T1 - Bayesian inference of scaled versus fractional Brownian motion JF - Journal of physics : A, mathematical and theoretical N2 - We present a Bayesian inference scheme for scaled Brownian motion, and investigate its performance on synthetic data for parameter estimation and model selection in a combined inference with fractional Brownian motion. We include the possibility of measurement noise in both models. We find that for trajectories of a few hundred time points the procedure is able to resolve well the true model and parameters. Using the prior of the synthetic data generation process also for the inference, the approach is optimal based on decision theory. We include a comparison with inference using a prior different from the data generating one. KW - Bayesian inference KW - scaled Brownian motion KW - single particle tracking Y1 - 2022 U6 - https://doi.org/10.1088/1751-8121/ac60e7 SN - 1751-8113 SN - 1751-8121 VL - 55 IS - 19 PB - IOP Publ. Ltd. CY - Bristol ER - TY - JOUR A1 - Cherstvy, Andrey G. A1 - Safdari, Hadiseh A1 - Metzler, Ralf T1 - Anomalous diffusion, nonergodicity, and ageing for exponentially and logarithmically time-dependent diffusivity BT - striking differences for massive versus massless particles JF - Journal of physics. D, Applied physics N2 - We investigate a diffusion process with a time-dependent diffusion coefficient, both exponentially increasing and decreasing in time, D(t)=D-0(e +/- 2 alpha t). For this (hypothetical) nonstationary diffusion process we compute-both analytically and from extensive stochastic simulations-the behavior of the ensemble- and time-averaged mean-squared displacements (MSDs) of the particles, both in the over- and underdamped limits. Simple asymptotic relations derived for the short- and long-time behaviors are shown to be in excellent agreement with the results of simulations. The diffusive characteristics in the presence of ageing are also considered, with dramatic differences of the over- versus underdamped regime. Our results for D(t)=D-0(e +/- 2 alpha t) extend and generalize the class of diffusive systems obeying scaled Brownian motion featuring a power-law-like variation of the diffusivity with time, D(t) similar to t(alpha-1). We also examine the logarithmically increasing diffusivity, D(t)=D(0)log[t/tau(0)], as another fundamental functional dependence (in addition to the power-law and exponential) and as an example of diffusivity slowly varying in time. One of the main conclusions is that the behavior of the massive particles is predominantly ergodic, while weak ergodicity breaking is repeatedly found for the time-dependent diffusion of the massless particles at short times. The latter manifests itself in the nonequivalence of the (both nonaged and aged) MSD and the mean time-averaged MSD. The current findings are potentially applicable to a class of physical systems out of thermal equilibrium where a rapid increase or decrease of the particles' diffusivity is inherently realized. One biological system potentially featuring all three types of time-dependent diffusion (power-law-like, exponential, and logarithmic) is water diffusion in the brain tissues, as we thoroughly discuss in the end. KW - anomalous diffusion KW - scaled Brownian motion KW - stochastic processes KW - nonstationary diffusivity KW - water diffusion in the brain KW - nonergodicity Y1 - 2021 U6 - https://doi.org/10.1088/1361-6463/abdff0 SN - 0022-3727 SN - 1361-6463 VL - 54 IS - 19 PB - IOP Publ. Ltd. CY - Bristol ER - TY - JOUR A1 - Safdari, Hadiseh A1 - Cherstvy, Andrey G. A1 - Chechkin, Aleksei V. A1 - Thiel, Felix A1 - Sokolov, Igor M. A1 - Metzler, Ralf T1 - Quantifying the non-ergodicity of scaled Brownian motion JF - Journal of physics : A, Mathematical and theoretical N2 - We examine the non-ergodic properties of scaled Brownian motion (SBM), a non-stationary stochastic process with a time dependent diffusivity of the form D(t) similar or equal to t(alpha-1). We compute the ergodicity breaking parameter EB in the entire range of scaling exponents a, both analytically and via extensive computer simulations of the stochastic Langevin equation. We demonstrate that in the limit of long trajectory lengths T and short lag times Delta the EB parameter as function of the scaling exponent a has no divergence at alpha - 1/2 and present the asymptotes for EB in different limits. We generalize the analytical and simulations results for the time averaged and ergodic properties of SBM in the presence of ageing, that is, when the observation of the system starts only a finite time span after its initiation. The approach developed here for the calculation of the higher time averaged moments of the particle displacement can be applied to derive the ergodic properties of other stochastic processes such as fractional Brownian motion. KW - scaled Brownian motion KW - anomalous diffusion KW - ageing Y1 - 2015 U6 - https://doi.org/10.1088/1751-8113/48/37/375002 SN - 1751-8113 SN - 1751-8121 VL - 48 IS - 37 PB - IOP Publ. Ltd. CY - Bristol ER -