Detail Essays in macroeconomic forecasting Bibliografi Author: Smith, Gregor W. (Advisor); Yetman, James Arthur ; Gregory, Allan W. (Advisor) Topik: ECONOMICS; GENERAL Bahasa: (EN )    ISBN: 0-612-35986-7     Penerbit: QUEEN'S UNIVERSITY AT KINGSTON (CANADA)     Tahun Terbit: 1998     Jenis: Theses - Dissertation Fulltext: NQ35986.pdf (0.0B; 0 download) Abstract A panel of forecasts may be defined to be in consensus when individual forecasters place identical weights on a common latent variable. Using this definition, a dynamic latent-variable model is formulated to test for consensus. This method also tests whether it is valid to use the mean forecast as the consensus forecast. In applications to surveys of U.S. macroeconomic forecasters, there is greater consensus in forecasts of unemployment than inflation, but idiosyncratic forecast autocorrelation is present from year to year for most forecasters. There is also evidence of greater consensus at shorter forecast horizons. When professional forecasters repeatedly forecast the same variable, their forecasts converge over time. That is, the differences between forecasters present in a panel decay with repeated revisions. This convergence is analyzed using a parametric decay function which permits flexibility in the decay path. The evidence suggests that unemployment forecast differences decay to zero at the same point in time for all forecasters, while decay in output forecasts is uneven across forecasters. Possible explanations for this, and their implications for studies using panels of forecasters, are discussed. Testing for consensus and modeling forecast convergence implicitly incorporates the idea that forecasters influence each other as they repeatedly forecast the same variable. Finally, this interaction is modeled explicitly. A forecaster loss function is estimated in which forecasters are assumed to trade off the distance of their forecast from competing predictions with the distance of their forecast from the actual outcome. Lemmings are defined as those forecasters who seek to be close to the forecasts of others at the expense of accuracy, while leaders are those who seek to distinguish their forecasts from others. A simple test is used to determine how these factors differ across forecasters. The results indicate that forecasters seek to distinguish their forecasts from those of competing forecasters. Opini Anda Klik untuk menuliskan opini Anda tentang koleksi ini! Lihat Sejarah Pengadaan  Konversi Metadata   Kembali

Copyright © 2006, 2007 Unika Atma Jaya, all rights reserved