Model Risk Cheat Sheet (DRAFT) by AshRoy

Model risk arises due to several factors: Assump­tions: Models rely on certain assump­tions, and if these assump­tions are inaccurate or unreal­istic, the model's output may not reflect the actual situation. Data Quality: The accuracy and comple­teness of input data signif­icantly impact the reliab­ility of model outputs. Poor data quality can lead to flawed results. Model Comple­xity: As models become more complex to capture real-world scenarios, their inherent uncert­ainty and vulner­ability to errors increase. Model Develo­pment and Implem­ent­ation: Errors or omissions during the model develo­pment and implem­ent­ation phases can introduce biases or inaccu­racies. Model Valida­tion: If the model validation process is inadequate or not indepe­ndent, it may fail to identify potential weaknesses or limita­tions. Behavioral Assump­tions: Many models assume that human behavior follows specific patterns, but real-world behavior may deviate from these assump­tions.