Search results

Bayesian additive regression trees (BART) is a fully Bayesian approach to modeling with ensembles of trees. BART can uncover complex regression functions with high-dimensional regressors in a fairly automatic way and provide Bayesian quantification of the uncertainty through the posterior. However, BART assumes independent and identical distributed (i.i.d) normal errors. This strong parametric assumption can lead to misleading inference and uncertainty quantification. In this chapter we use the classic Dirichlet process mixture (DPM) mechanism to nonparametrically model the error distribution. A key strength of BART is that default prior settings work reasonably well in a variety of problems. The challenge in extending BART is to choose the parameters of the DPM so that the strengths of the standard BART approach is not lost when the errors are close to normal, but the DPM has the ability to adapt to non-normal errors.

For nearly 80 years, the field of macroeconomics has largely been shaped by the aftermath of the Keynesian revolution. Many economists have argued that this revolution and the subsequent internal and external disputes it has sparked have had the unfortunate side effect of crowding out much of what was good in macro-level analysis before it, leading to the dissatisfactory state of macroeconomics we have today. In the search for alternative paths for macroeconomics, I focus on two separate but compatible traditions: monetary disequilibrium (MD) theory and the Austrian business cycle theory (ABCT). I argue that scholars in these traditions employed a far richer micro-theoretic explanation for the business cycle well before Keynes’s General Theory. Unfortunately, their ideas were not united in time to mount a sufficient counterattack to the Keynesian crusade. My goal is to unite the best elements of these two traditions by providing what I believe is the “missing link” that can help connect these alternative paths: free banking theory.

Norbert Wiener and Warren S. McCulloch are presented as the main contemporary initiators of cybernetics. Wiener chose Leibniz as “patron saint” of cybernetics because of his works on general logics and binary arithmetics. On the contrary McCulloch chose Descartes, having observed the intervention of a feedback in his conceptions about perception and consecutive action. The two paradigmatic views of Wiener and McCulloch may indicate two streams of evolution for cybernetics, leading, on the one side, to general views about perception and action or “epistemo‐praxiology” and, on the other side, to considerations about interacting systems such as a generalized world‐wide network comparable to a “creature of gigantic dimensions” covering the globe.

Elie Halévy essentially expressed the view recorded by James Mill in his anonymously written ‘On the Nature, Measures, and Causes of Value’⁷ that the first chapter of the Critical Dissertation relating to the nature of value ‘contains not an assertion, who which, as far as ideas politico-economical are concerned, Mr. Ricardo would not have assented; it contains, not indeed, as far as such ideas are concerned, an assertion which is not implied in the propositions which Mr. Ricardo has put forth. It is a criticism on some of Mr. Ricardo's forms of expression…’ ([J. Mill], 1826a, p. 157). The justification for the Ricardian reaction is clear enough, as I shall now show.⁸

Though contemporaries, Adam Smith and Sir James Steuart are commonly portrayed as if they belonged to different eras. Whereas Smith went down in history as both founder of the science of political economy and patron saint of economic liberalism, Steuart became known as the last, outdated advocate for mercantilist policies in Britain. Smith himself was responsible for popularizing the notion of the “system of commerce” as an approach to political economy that dominated the early modern period. As a historiographical concept, the mercantile system became a misguided international trade theory grounded upon the Midas fallacy and the favorable balance of trade doctrine. Smith’s treatment of international trade in the Wealth of Nations, however, was criticized for its inconsistencies and lack of analytical clarity even by some among his own followers. Given Smith’s doubtful credentials as an international trade theorist, the chapter investigates the reasons that led him and Steuart to be placed on opposite sides of the mercantilist divide. The authors analyze the works of both authors in depth, showing that their disagreements had chiefly to do with different views on money and monetary policy. Additionally, the authors explore how early nineteenth-century writers such as Jean-Baptiste Say and J. R. McCulloch helped forge the intellectual profiles of both Steuart and Smith.

The founders of cybernetics and systems are presented, among them N. Wiener, W.S. McCulloch and L. von Bertalanffy. Some precursors are cited from antiquity to 20th century. The basic concepts are exposed: feedback, quantity of information, requisite variety, homeostasis, local and global points of view, oprn systems, autopoiesis. The roles of the observer and of the actor are emphasized. Future is considered in three directions: development of epistemology and of praxiology, symbiosis of man and machine, role of requisite variety in the survival of mankind.