The use of agent-based models (ABMs) has increased in the last years to simulate social systems and, in particular, financial markets. ABMs of financial markets are usually validated by checking the ability of the model to reproduce a set of empirical stylised facts. However, other common-sense evidence is available which is often not taken into account, ending with models which are valid but not sensible. In this paper we present an ABM of a stock market which incorporates this type of common-sense evidence and implements realistic trading strategies based on practitioners literature. We next validate the model using a comprehensive approach consisting of four steps: assessment of face validity, sensitivity analysis, calibration and validation of model outputs.

中文翻译：

---

在基于代理的股票市场模型中使用现实的交易策略

近年来，基于代理的模型（ABM）的使用在增加，以模拟社会系统，尤其是金融市场。金融市场的ABM通常通过检查模型再现一组经验化的程式化事实的能力来验证。但是，可以得到其他常识性证据，而这些常识性证据通常没有被考虑在内，但以有效但不明智的模型结尾。在本文中，我们介绍了一种股票市场的ABM，其中包含了这种常识性证据，并根据从业人员文献实施了现实的交易策略。接下来，我们将使用包括以下四个步骤的综合方法来验证模型：面部有效性评估，灵敏度分析，模型输出的校准和验证。