Behavior is endogenous forumulation of nonphyiscal stocks. However, with heterogenity added to this, different behaviors of the people can be explained which is explained by strong prior that differs in groups of people. Both physical categories (vaccinated or not) to nonphysical (political party, region) can affect this prior formation. More concrete prior formulation is in [[FinanceBehavior]] which can be connected to vaccination of covid in 'risk perception' context. > Bayesian statisticians are often wary of strong priors because they want to ‘let the data speak’. The trouble is that letting the data speak in this way will result in many erroneously strong statements. Paradoxically, weak priors imply inappropriately strong conclusions in certain dimensions of the posterior, a point which we have pointed out before in more complex multivariate examples; see section 3 of Gelman (1996). Again, our point is not that a high-variance prior for a parameter of interest is necessarily wrong; rather we are emphasizing that using such a prior has implications for the applicability of the model, as it can result in strong posterior claims from what might seem like weak evidence. Q1. Can I connect this with path dependency? Q2. How can this portrayal (elicitation and its mechanism to affect decision making) be embedded in Bayesian workflow in the most automated way? Q3. Prior elicitation of GAN? Starting from different groups of prior and if that passes the test (decieves the discriminator), that can serve as the test for behavior (i.e. good enough number of classification and its generating process details). Q4. #hr says "behaviorally realistic potrayal of decision making is at the heart of system dynamics". #ag says in his paper, > incoherence of Bayesian updating eliminates any theoretical basis for a behavioristic interpretation of the prior distribution; > ... > Bayesian inference goes from likelihood and prior to posterior, a deterministic procedure followed by Bayes’ theorem after one model is set. Bayesian logic contains both Bayesian inference and decision theory. Bayesian data analysis involves three steps of model building, inference, and model checking and improvement. A complete Bayesian workflow refers to Bayesian data analysis for a sequence of models, not just for the purpose of model choice or model averaging but more importantly to understand these models. From this sense, Path dependency and increasing return concept is introduced: > we routinely assume a joint distribution and we routinely treat the act of measurement as a direct application of conditional probability. If classical probability theory needs to be generalized to apply to quantum mechanics, then it makes us wonder if it should be generalized for applications in political science, economics, psychometrics, astronomy, and so forth. > ... > It is not clear if there are any practical uses to this idea in statistics, outside of quantum physics. For example, would it make sense to use ‘two-slit-type’ models in psychometrics, to capture the idea that asking one question affects the response to others? Would it make sense to model macroscopic phenomena in the physical, biological, and social sciences using complex amplitudes—thus incorporating ideas of phase and coherence into applied Bayesian inference? Could nodes,wave behavior,entanglement,and other quantumphenomenamanifest in observable ways in applied statistics, motivating models that go beyond classical Boltzmann probabilities? We have no idea.