The DARC Toolbox: automated, flexible, and efficient delayed and risky choice experiments using Bayesian adaptive design
By: Researcher
Disclaimer: The provided code links for this paper are external links. Science Nest has no responsibility for the accuracy, legality or content of these links. Also, by downloading this code(s), you agree to comply with the terms of use as set out by the author(s) of the code(s).
Please contact us in case of a broken link from here
| Authors | Benjamin T. Vincent, Thomas Rainforth |
| Journal/Conference Name | PsyArXiv Preprints |
| Paper Category | Economics and Econometrics |
| Paper Abstract | Delayed and risky choice (DARC) experiments are a cornerstone of research in psychol-ogy, behavioural economics and neuroeconomics. By collecting an agent’s preferencesbetween pairs of prospects we can characterise their preferences, investigate what affectsthem, and probe the underlying decision making mechanisms. We present a state-of-the-artapproach and software toolbox allowing such DARC experiments to be run in a highly effi-cient way. Data collection is costly, so our toolbox automatically and adaptively generatespairs of prospects in real time to maximise the information gathered about the participant’sbehaviours. We demonstrate that this leads to improvements over alternative experimen-tal paradigms. The key to realising our real time and automatic performance is a numberof advances over current Bayesian adaptive design methodology. In particular, we derivean improved estimator for discrete output problems and design a novel algorithm for au-tomating sequential adaptive design. We provide a number of pre-prepared DARC toolsfor researchers to use, but a key contribution is an adaptive experiment toolbox that canbe extended to virtually any 2-alternative-choice task. In particular, to carry out customadaptive experiments using our toolbox, the user need only encode their behavioural modeland design space – both the subsequent inference and sequential design optimisation areautomated for arbitrary models the user might write. |
| Date of publication | 2017 |
| Code Programming Language | Matlab |
| Comment |
Copyright Researcher 2021