Collection and Analysis of the Data for an elucidation of the stock price mechanism

在这三年里，我每年进行两次调查，分别在8月和2月。共向银行、证券公司、投资信托公司、投资咨询公司等约400家机构投资者发放了约40个问题的调查问卷。回复率约为35% ~ 38%。在这次调查中，我问了他们对日本和美国未来股价的预期、对当前市场状况的评价、对未来地价的预测等问题。每次的回应都被总结成一份10页的报告。这项调查是1989年以来进行的调查的一部分。美国耶鲁大学的席勒教授提出了一些主要问题。日本和美国投资者的答案的比较给了我们非常有趣的知识，我们写了一篇论文，阐明了1990年以来日本股票价格崩溃的原因（经济与统计评论，1996年2月）。我写了一篇分析美国股票价格现状的文章（日经日报，1997年10月）。他说：“我认为美国股票价格处于泡沫状态，但泡沫比日本80年代末的泡沫要小。我还指出，美国投资者比上世纪80年代末的日本投资者更冷静。现在，我要检验投资者的预期是否为理性预期。以前的许多研究都认为，调查所得的预期数据是不理性的，有一篇论文认为，如果我们承认一部分市场参与者是图表主义者，就可以填补与理性预期的偏差。由于我们询问受访者是原教旨主义者还是图表主义者，因此我们可以使用我们的调查数据明确该假设是否有效；这就是我们现在要做的。

I conducted a survey twice a year, in August and February for these three years. Questionnaires that consist of about 40 questions were sent to about 400 institutional investors, i.e.banks, securities companies, investment trust, and investment consultant companies. The response rate was about 35 to 38%. In this survey, I asked their expectations on the future stoch prices in Japan and the U.S., evaluation of the current market conditions, predictions about future land prices, and so on. Responses are summarized in a ten-page report every time.This survey is a part of those conducted since 1989. Main questions have been asked in the U.S.by Professor Shiller of Yale University. A comparison of the Japanese and U.S.investors' answers gives us very interesting knowledge, and we wrote a paper elucidating the reason of the crash of the Japanese stock prices since 1990 (Review of Economics and Statistics, February 1996).I wrote an article analyzing the current situation of the U.S.stock prices (The Nikkei Daily Newspaper, October 1997). I judge that the U.S.stock prices are in a bubble situation, but the bubble is smaller than the one in the late 1980s in Japan. I also point out there that the U.S.investors are cooler than the Japanese investors in the late 1980s.Now, I am going to test whether or not the investors' expectations are rational expectation. Many previous studies argued that expectation data from surveys is not rational, and one paper argued that the deviation from rational expectation could be filled if we admit that a part of market participants is chartist. Since we asked our respondents if they are fundamentalist or chartist, we can make clear whether or not this hypothesis is valid, using our survey data ; which is now we are going to do.