Author
Listed:
- Njiru, Ruth
- Appel, Franziska
- Dong, Changxing
- Balmann, Alfons
Abstract
In light of the dynamic challenges facing agricultural land markets, the conventional analytical frameworks fall short in capturing the intricate interplay of strategic decisions and evolving complexities. This necessitates the development of a novel method, integrating deep learning into Agent-based Modelling, to provide a more realistic and nuanced understanding of land market dynamics, enabling informed policy assessments and contributing to a comprehensive discourse on agricultural structural change. In this paper, different deep learning models are tested and evaluated, as emulators of AgriPoliS (Agricultural Policy Simulator). AgriPoliS is an agent-based model used to model the evolution of structural change in agriculture resultant on the change in the policy environment. This study is part of preliminary works towards integrating deep learning methods and predictions with AgriPoliS to capture strategic decision making and actions of agents in land markets. The paper tests the models on their suitability, computational requirements and run-time complexities. The output from AgriPoliS serves as the input features for the deep learning models. Models are evaluated using a combination of coefficient of determination (R2 score), mean absolute error, visual displays and runtime. The models were able to replicate the variable of interest with a high degree of accuracy with R2 score of more than 90%. The CNN was the most suited for replicating the data. Through this work, we learned the required complexities, computational and training efforts needed to integrate deep learning and AgriPoliS to capture strategic decision-making.
Suggested Citation
Njiru, Ruth & Appel, Franziska & Dong, Changxing & Balmann, Alfons, 2024.
"Application of Deep Learning to Emulate an Agent-Based Model,"
FORLand Project Publications
340874, University of Natural Resources and Applied Life Sciences, Vienna, Department of Economics and Social Sciences.
Handle:
RePEc:ags:bokufo:340874
DOI: 10.22004/ag.econ.340874
Download full text from publisher
Corrections
All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:ags:bokufo:340874. See general information about how to correct material in RePEc.
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
We have no bibliographic references for this item. You can help adding them by using this form .
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: AgEcon Search (email available below). General contact details of provider: https://edirc.repec.org/data/wbokuat.html .
Please note that corrections may take a couple of weeks to filter through
the various RePEc services.