A Time-Series Study of the Formation and Predictive Performance of EEC Production Survey Expectations
分析欧洲五国制造业调查中的短期生产预期，发现预期能有效利用已有信息并预测企业自报产量，但无法预测官方统计的客观产量，说明其价值取决于产量衡量方式。
Abstract This article provides an empirical analysis of manufacturers' survey responses on short-term production expectations. If such surveys provide valid anticipations of production activity in customer industries, they should be valuable information to management in supplier industries. Two major questions are considered: (a) Are survey expectations valid predictors of future production activity as reported later by the manufacturers and/or as measured by official statistics? (b) Do the expectations make efficient use of information available to the surveyed manufacturers? The research design involves univariate and multivariate time-series analysis of monthly data for five European countries in three major sectors of the manufacturing industry, using the concept of Granger causality. The main findings are that the survey expectations often make an efficient use of the information available to the surveyed manufacturers and that the anticipations Granger-cause survey-reported production levels, but that they do not Granger-cause objectively measured production levels. These results suggest that the value of such survey expectations is contingent on the way in which manufacturing activity is measured.