Rational herds: Economic models of social learning by Christophe P. Chamley

By Christophe P. Chamley

Penguins leaping off a cliff, monetary forecasters and fiscal advisors speculating opposed to a forex, and farmers utilizing conventional tools in India are all practicing social studying. Such studying from the habit of others may possibly and does bring about herds, crashes, and booms. those matters became, over the past ten years, an exhilarating box of analysis in theoretical and utilized economics, finance, and in different social sciences. This 2004 e-book presents either a casual advent and in-depth insights into the topic. each one bankruptcy is dedicated to a separate factor: contributors study from the observations of activities, the results of those activities, and from what others say. they could hold up or make a right away choice; they could compete opposed to others or achieve from cooperation; they make judgements approximately funding, crop offerings, and monetary investments. The publication highlights the similarities and the diversities among many of the circumstances.

Show description

Read or Download Rational herds: Economic models of social learning PDF

Similar macroeconomics books

Currency competition and foreign exchange markets

Foreign money pageant and foreign currency Markets is an incredible new theoretical and empirical examine of foreign currencies that makes a speciality of the function the Euro (the destiny ecu foreign money) will play within the overseas financial and monetary process, in addition to the united states greenback and the japanese yen. not like a lot of the prevailing literature that techniques the topic from a macroeconomic point of view, Philipp Hartmann develops a theoretical version that makes use of online game concept, time sequence and panel econometrics, and hyperlinks monetary markets research with transaction expense economics.

Money, Crises, and Transition: Essays in Honor of Guillermo A. Calvo

Essays via renowned students and policymakers honor the most influential macroeconomists of the final thirty years, discussing the subjects at the back of his paintings.

Market Socialism or the Restoration of Capitalism? (International Council for Central and East European Studies)

The cave in of communism in jap Europe has raised a few questions in regards to the destiny process their economies. Has capitalism gained or is anything varied rising? Has marketplace socialism vanished for stable? How can the transitionary interval be controlled and what impact will it have at the way of life in jap Europe?

Handbook of Tourism Economics: Analysis, New Applications and Case Studies

Guide of Tourism Economics: research, New functions and Case reviews offers an up to date, concise and readable insurance of crucial subject matters in tourism economics. It can pay realization to suitable conventional issues in tourism economics in addition to interesting rising themes during this box -- themes that are anticipated to be of continuous significance.

Additional info for Rational herds: Economic models of social learning

Sample text

8 n A Heuristic Remark on the Proof of the Martingale Convergence Theorem The main intuition of the proof is important for our understanding of Bayesian learning. It is a formalization13 of the metaphor of the drunkard. In words, the definition of a martingale states that agents do not anticipate systematic errors. This implies that the updating difference µt+1 − µt is uncorrelated with µt . The same property holds for more distant periods: conditional on the information in period t, the random variables µt+k+1 − µt+k are uncorrelated for k ≥ 0.

These principles hold for a wider class of models, but only the Gaussian model provides such a simple formulation: 1. The normal distribution is summarized by the most intuitive two parameters of a distribution, the mean and the variance (or its reciprocal, the precision). 2. The updating rules for both the mean and the precision are linear. This makes any computation easier. 3. The weight of the private signal s depends on the noise-to-signal ratio in the most intuitive way. When the variance of the noise term, σ 2 , tends to zero, or equivalently its precision tends to infinity, the signal’s weight α tends to one and the weight of the ex ante expected value of θ tends to zero.

An agent has an incentive to change less than he would with no evaluation, because he wants to show that he already knows much (that his ρ is high). Prendergast and Stole (1996) analyze the distortion that is created by the evaluation of the agent’s actions. 5), the expected value of θ at the end of period t is N (µt+1 , ρt+1 ) with µt+1 = ρ ρθ + (t − 1)ρ st + µt . 5 Appendix 41 The weight on the new signal s t increases with ρ. Note, however, that the weight does not tend to 1 as ρ → ∞ (as it would in the first period), because a high ρ means that previous estimates of θ have also a high precision (because they are made with signals of high precision).

Download PDF sample

Rated 4.47 of 5 – based on 24 votes