nep-for New Economics Papers
on Forecasting
Issue of 2019‒05‒13
four papers chosen by
Rob J Hyndman
Monash University

  1. High-Frequency Credit Spread Information and Macroeconomic Forecast Revision By Bruno Deschamps; Christos Ioannidis; Kook Ka
  2. Statistical Learning and Exchange Rate Forecasting By Emilio Colombo; Matteo Pelagatti
  3. Assessing financial stability risks from the real estate market in Italy: an update By Federica Ciocchetta; Wanda Cornacchia
  4. Forecasting Japanese inflation with a news-based leading indicator of economic activities By Keiichi Goshima; Hiroshi Ishijima; Mototsugu Shintani; Hiroki Yamamoto

  1. By: Bruno Deschamps (Nottingham University Business School China); Christos Ioannidis (Aston Business School, Aston University); Kook Ka (Economic Research Institute, Bank of Korea)
    Abstract: We examine whether professional forecasters incorporate high-frequency information about credit conditions when revising their economic forecasts. Using Mixed Data Sampling regression approach, we find that daily credit spreads have significant predictive ability for monthly forecast revisions of output growth, at both aggregate and individual forecast levels. The relations are shown to be notably strong during ¡®bad¡¯ economic conditions, suggesting that forecasters anticipate more pronounced effects of credit tightening during economic downturns, indicating the amplification effect of financial developments on macroeconomic aggregates. Forecasts do not incorporate the totality of financial information received in equal measures, implying the presence of information rigidities in the incorporation of credit spread information.
    Keywords: Forecast Revision, GDP Forecast, Credit Spread, High-Frequency Data, Mixed Data Sampling (MIDAS)
    JEL: C53 E32 E44
    Date: 2019–05–03
  2. By: Emilio Colombo; Matteo Pelagatti
    Abstract: his study uses the most innovative tools recently proposed in the statistical learning literature to assess the ability of standard exchange rate models to predict the exchange rate in the short and long run. Our results show that statistical learning methods display impressive performances, consistently outperforming the random walk in forecasting the exchange rate at different forecasting horizons, with the exception of the very short term (a period of 1-2 months). We use these tools to compare the predictive ability of different exchange rate models and model specifications. We find that sticky price versions of the monetary model with the error correction specification exhibit the best performance. We also explore the functioning of statistical learning models by developing measures of variable importance and by analyzing the kind of relationship that links each variable with the outcome. This allows us to improve our understanding of the relationship between the exchange rate and economic fundamentals, which appears complex and characterized by strong non-linearities.
    JEL: F37 C53
    Date: 2019
  3. By: Federica Ciocchetta (Bank of Italy); Wanda Cornacchia (Bank of Italy)
    Abstract: We provide an update of the analytical framework to assess financial stability risks arising from the real estate sector in Italy. The enhancement concerns the definition of a new vulnerability indicator, measured in terms of the flow of total non-performing loans (NPLs) and not, as done previously, in terms of bad loans only. We focus separately on households (as an approximation for residential real estate, RRE) and on firms engaged in construction, management and investment services in the real estate sector (as an approximation for commercial real estate, CRE). Two early warning models are estimated using the new vulnerability indicator for RRE and CRE, respectively, as dependent variable. Both models exhibit good forecasting performances: the median predictions fit well the new vulnerability indicators in out-of-sample forecasts. Overall, models’ projections indicate that potential risks for banks stemming from the real estate sector will remain contained in the next few quarters.
    Keywords: real estate markets, early warning models, bayesian model averaging, banking crises
    JEL: C52 E58 G21
    Date: 2019–04
  4. By: Keiichi Goshima (Waseda University and Bank of Japan); Hiroshi Ishijima (Chuo University); Mototsugu Shintani (Corresponding author, The University of Tokyo and Bank of Japan); Hiroki Yamamoto (The University of Tokyo and Bank of Japan)
    Abstract: We construct business cycle indexes based on the daily Japanese newspaper articles and estimate the Phillips curve model to forecast inflation at a daily frequency. We find that the news-based leading indicator, constructed from the topic on future economic conditions, is useful in forecasting the inflation rate in Japan.
    Date: 2019–05

This nep-for issue is ©2019 by Rob J Hyndman. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.