Parameter estimation from aggregate observations: a Wasserstein distance-based sequential Monte Carlo sampler
In this work, we study systems consisting of a group of moving particles. In such systems, often some important parameters are unknown and have to be estimated from observed data. Such parameter estimation problems can often be solved via a Bayesian inference framework. However, in many practical pr...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The Royal Society
2023-08-01
|
| Series: | Royal Society Open Science |
| Subjects: | |
| Online Access: | https://royalsocietypublishing.org/doi/10.1098/rsos.230275 |
| _version_ | 1797749123432054784 |
|---|---|
| author | Chen Cheng Linjie Wen Jinglai Li |
| author_facet | Chen Cheng Linjie Wen Jinglai Li |
| author_sort | Chen Cheng |
| collection | DOAJ |
| description | In this work, we study systems consisting of a group of moving particles. In such systems, often some important parameters are unknown and have to be estimated from observed data. Such parameter estimation problems can often be solved via a Bayesian inference framework. However, in many practical problems, only data at the aggregate level is available and as a result the likelihood function is not available, which poses a challenge for Bayesian methods. In particular, we consider the situation where the distributions of the particles are observed. We propose a Wasserstein distance (WD)-based sequential Monte Carlo sampler to solve the problem: the WD is used to measure the similarity between the observed and the simulated particle distributions and the sequential Monte Carlo samplers is used to deal with the sequentially available observations. Two real-world examples are provided to demonstrate the performance of the proposed method. |
| first_indexed | 2024-03-12T16:14:40Z |
| format | Article |
| id | doaj.art-a08b72839cac4124a9acbe9eeeb246b9 |
| institution | Directory Open Access Journal |
| issn | 2054-5703 |
| language | English |
| last_indexed | 2024-03-12T16:14:40Z |
| publishDate | 2023-08-01 |
| publisher | The Royal Society |
| record_format | Article |
| series | Royal Society Open Science |
| spelling | doaj.art-a08b72839cac4124a9acbe9eeeb246b92023-08-09T07:05:25ZengThe Royal SocietyRoyal Society Open Science2054-57032023-08-0110810.1098/rsos.230275Parameter estimation from aggregate observations: a Wasserstein distance-based sequential Monte Carlo samplerChen Cheng0Linjie Wen1Jinglai Li2School of Mathematical Sciences, Shanghai Jiao Tong University, Shanghai 200240, People’s Republic of ChinaSchool of Earth and Space Sciences, Peking University, 5 Yiheyuan Rd, Beijing 100871, People’s Republic of ChinaSchool of Mathematics, University of Birmingham, Birmingham B15 2TT, UKIn this work, we study systems consisting of a group of moving particles. In such systems, often some important parameters are unknown and have to be estimated from observed data. Such parameter estimation problems can often be solved via a Bayesian inference framework. However, in many practical problems, only data at the aggregate level is available and as a result the likelihood function is not available, which poses a challenge for Bayesian methods. In particular, we consider the situation where the distributions of the particles are observed. We propose a Wasserstein distance (WD)-based sequential Monte Carlo sampler to solve the problem: the WD is used to measure the similarity between the observed and the simulated particle distributions and the sequential Monte Carlo samplers is used to deal with the sequentially available observations. Two real-world examples are provided to demonstrate the performance of the proposed method.https://royalsocietypublishing.org/doi/10.1098/rsos.230275parameter estimationsequential Monte Carlo samplerlikelihood-free inferenceWasserstein distance |
| spellingShingle | Chen Cheng Linjie Wen Jinglai Li Parameter estimation from aggregate observations: a Wasserstein distance-based sequential Monte Carlo sampler Royal Society Open Science parameter estimation sequential Monte Carlo sampler likelihood-free inference Wasserstein distance |
| title | Parameter estimation from aggregate observations: a Wasserstein distance-based sequential Monte Carlo sampler |
| title_full | Parameter estimation from aggregate observations: a Wasserstein distance-based sequential Monte Carlo sampler |
| title_fullStr | Parameter estimation from aggregate observations: a Wasserstein distance-based sequential Monte Carlo sampler |
| title_full_unstemmed | Parameter estimation from aggregate observations: a Wasserstein distance-based sequential Monte Carlo sampler |
| title_short | Parameter estimation from aggregate observations: a Wasserstein distance-based sequential Monte Carlo sampler |
| title_sort | parameter estimation from aggregate observations a wasserstein distance based sequential monte carlo sampler |
| topic | parameter estimation sequential Monte Carlo sampler likelihood-free inference Wasserstein distance |
| url | https://royalsocietypublishing.org/doi/10.1098/rsos.230275 |
| work_keys_str_mv | AT chencheng parameterestimationfromaggregateobservationsawassersteindistancebasedsequentialmontecarlosampler AT linjiewen parameterestimationfromaggregateobservationsawassersteindistancebasedsequentialmontecarlosampler AT jinglaili parameterestimationfromaggregateobservationsawassersteindistancebasedsequentialmontecarlosampler |