Constructing reality with models

Contextualist model evaluation: models in financial economics and index funds

“…Boldyrev and Ushakov ( 2016 ) argue that models can be built with the purpose of shaping the world and that this ought to be accounted for in our philosophical accounts of models: “We would propose to complement representational and non-representational accounts by looking at the ways economic modeling is aimed at transforming its own target.” (p. 41). By the same token, Tee ( 2019 ) argues against the traditional view in philosophy of modelling to take models as having only a “passive epistemic role”. Tee suggests that, in addition to the traditional criterion of model-world representation, at least some models could be evaluated by their “constructive” capacity.…”

“…Boldyrev and Svetlova ( 2016 ). Other recent work is Boldyrev and Ushakov ( 2016 ), who argue that models can be built with the purpose of shaping the world, as in the case of the models of general equilibrium built by Hurwicz, and that this ought to be accounted for in our philosophical accounts of models; Tee ( 2019 ) who argues that, in addition to the traditional criterion of model-world representation, at least some models could be evaluated by their “constructive” capacity; and van Basshuysen et al ( 2021 ), who, in the context of the Covid-19 pandemic suggest ways in which predictive epidemiological models affect their targets and raises important questions about the implications this has for accounts of models.…”

The modal basis of scientific modelling

“…Fictional models may seem to resist the type of de-idealisation that we have just discussed, so it may be more difficult to see how such models could have actual targets. The role of fictions in modelling has received plenty of attention (already in Cartwright, 1983;and also, e.g., Frigg, 2010;Bokulich, 2011;Tee, 2019;Kimpton-Nye, 2020). Let me continue with Bokulich, who outlines an example of an explanatory fictional model: Bohr's atom.…”

Managing Performative Models

“…Even if such efforts were successful, however, there is a second important problem with the appraisal view that we anticipated in van Basshuysen et al (2021) and that Winsberg and Harvard further discuss. To appreciate this problem, let me cast the appraisal view in somewhat clearer outlines: Appraisal: The overall goodness of a model (e.g., in terms of a wide understanding of adequacy-for-purpose ; see Parker 2020; van Basshuysen 2022), is a function of (1) whether a model properly performs its epistemic functions, e.g., issuing accurate predictions, providing adequate explanations or facilitating understanding of a phenomenon, and (2) whether a model contributes to the achievement of the practical purposes for which the model was constructed, including by causally affecting desired kinds of change in a target system (see Tee 2019). …”

“…Scientific models can be performative: in addition to serving various epistemic purposes, they can also causally affect phenomena, such as when agents’ behaviors change in response to model predictions. In recent years, philosophers have made substantial progress in delineating different forms of performativity and characterizing the problems they can pose, such as when the forecasts researchers derive from models are self-defeating and compromise models’ epistemic functioning (Avery et al 2020b; Godman and Marchionni 2022; Jiménez-Buedo 2021; Northcott 2022; Tee 2019; van Basshuysen 2022; van Basshuysen et al 2021; Vergara-Fernández, Heilmann, and Szymanowska 2023; Winsberg and Harvard 2022).…”