Detail Forecasting inflation with probit and regression models Bibliografi Author: Ratti, Ronald A. (Advisor); Kang, Sungjun Topik: ECONOMICS; GENERAL Bahasa: (EN )    ISBN: 0-599-48005-X     Penerbit: UNIVERSITY OF MISSOURI - COLUMBIA     Tahun Terbit: 1999     Jenis: Theses - Dissertation Fulltext: 9946268.pdf (0.0B; 2 download) Abstract This dissertation examines the predictive power of inflation indicators with a probit model. Until the late 1980s commodity price indexes and oil prices gave a good performance in forecasting inflation target ranges. These variables showed power in signaling the occurrence of high inflation. However, in the 1990s these variables show a sharp decrease in the ability to forecast inflation. Interest rates exhibit stable forecasting power for inflation, and if anything, show a better performance in the 1990s than in earlier period. In contrast to Cecchetti's (1995) regression model, in the probit model inflation indicators provide additional information over and above that provided by past inflation for point estimates for inflation and for inflation target ranges. The forecasting performance of the probit model is compared to that of regression models by using the probit model to generate point-estimates for future inflation. The point estimate for inflation in the probit model is the inflation rate at which there is fifty-fifty chance that inflation will be either below or above that point. Inflation forecasts from the probit model are better than those from the regression model at the 1-year horizon and 1–2 year horizon. 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