摘要: T he rational expectations revolution in economics was born of the recognition that people hold for future economic variables are fundamental to their behavior. Thus, our (economists') models require (those who make up economy) models, which they use generate expectations. The key idea is collapse two into one: assume know (or behave as if know) true model describes economy. This allows economists construct simple and elegant economy, appealing theoretically can be studied without collecting any data about minds actors. problem with collapsing kinds one a gross oversimplification. Obviously, popular (the used by broad masses actors form expectations) not same those held economists. Once accepts difference, modelling cannot proceed on themselves. paper reports such collection effort using questionnaire survey methods, purpose understanding speculative markets. I will report here my research understand U.S. stock market crash October 1987; Fumiko Konya, Yoshiro Tsutsui undertook Japanese Karl Case recent real estate booms; John Pound periodic "hot" markets initial public offerings (IPO's) common stock.