Combining Econometrics and Agent-based Modelling
4. Model verification and validation
Verification and validation is a crucial part of the overall MPMAS modelling process. It ensures that model inputs based on the lottery as well as model outcomes correspond to reality. In fact, validation is considered one of the most important steps in any quantitative modelling approach (Kaiser and Messer, 2012), e.g. farm level decision-making models. A review of 48 bio-economic farm models by Janssen and van Ittersum (2007) however showed that of the reviewed applications only 23 were validated, four of these quantitatively. The MPMAS application used here is based on established decision-making, crop growth and innovation diffusion modelling methods, while the model’s coefficients are estimated based on comprehensive and reliable survey data. This approach guarantees the construct validity of the model. In the following section, the verification and validation applied to the model’s results are explained in more detail. It should be noted however, that due to the nature of the model and the data collection process, the simulated and observed data are not independent.
4.1 Verification of asset allocations
Verification in the context of MPMAS implies checking that the resources allocated to agents are consistent with the observed resources available to farmers.
In the study model, consistency is tested on six important assets: household size, liquidity, greenhouses owned, area under chrysanthemum cultivation, area under litchi orchards and area under rose cultivation. All of these variables used in the test are expressed as per household quantities. Using linear regression without a...
You are not authenticated to view the full text of this chapter or article.
This site requires a subscription or purchase to access the full text of books or journals.
Do you have any questions? Contact us.Or login to access all content.