nep-for New Economics Papers
on Forecasting
Issue of 2016‒11‒13
six papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting the Probability of Recessions in South Africa: The Role of Decomposed Term-Spread and Economic Policy Uncertainty By Goodness C. Aye; Christina Christou; Luis A. Gil-Alana; Rangan Gupta
  2. A dynamic factor model for forecasting house prices in Belgium By Marina Emiris
  3. Would DSGE Models have predicted the Great Recession in Austria? By Fritz Breuss
  4. Forecasting stock market returns by summing the frequency-decomposed parts By Gonçalo Faria; Fabio Verona
  5. What is the Expected Return on a Stock? By Martin, Ian; Wagner, Christian
  6. Simultaneous Ensemble Post-Processing for Multiple Lead Times with Standardized Anomalies By Markus Dabernig; Georg J. Mayr; Jakob W. Messner; Achim Zeileis

  1. By: Goodness C. Aye (Department of Economics, University of Pretoria, South Africa); Christina Christou (School of Economics and Management, Open University of Cyprus, Cyprus); Luis A. Gil-Alana (Universidad de Navarra, Faculty of Economics and Business Administration, Spain); Rangan Gupta (Department of Economics, University of Pretoria, South Africa)
    Abstract: This paper extends the vast literature forecasting the probability of recession by including the different components of the term spread, namely the expectation and the term premium components obtained from a fractional integration approach. We also augment these with the economic policy uncertainty index. We use 10 specifications of the probit prediction model and quarterly data from South Africa covering the period 1990:1 to 2012:1 for analyses. Our out-of-sample results show that the model that incorporates the expectation component of the yield spread in addition to economic policy uncertainty provides the best forecast of recession in South Africa. All three recession periods in our sample were accurately dictated by the prediction models and the best forecast occurred at the four quarters ahead horizon. These results were also robust to the full sample prediction
    Keywords: Expected term spread; term premium; economic policy uncertainty; recession; out-of-sample forecast; Probit model
    JEL: C25 E37 E44 E52 E62
    Date: 2016–11
  2. By: Marina Emiris (Economics and Research Department, National Bank of Belgium)
    Abstract: The paper forecasts the residential property price index in Belgium with a dynamic factor model (DFM) estimated with a dataset of macro-economic variables describing the Belgian and euro area economy. The model is validated with out-of-sample forecasts which are obtained recursively over an expanding window over the period 2000q1-2012q4. We illustrate how the model reads information from mortgage loans, interest rates, GDP and inflation to revise the residential property price forecast as a result of a change in assumptions for the future paths of these variables
    Keywords: dynamic factor model, conditional forecast, house prices
    JEL: E32 G21 C53
    Date: 2016–11
  3. By: Fritz Breuss (WIFO)
    Abstract: DSGE (Dynamic stochastic general equilibrium) models are the common workhorse of modern macroeconomic theory. Whereas story-telling and policy analysis were in the forefront of applications since its inception, the forecasting perspective of DSGE models is only recently topical. In this study, we perform a post-mortem analysis of the predictive power of DSGE models in the case of Austria's recession in 2009. For this purpose, 8 DSGE models with different characteristics (small and large models, closed and open economy models, one and two-country models) were used. The initial hypothesis was that DSGE models are inferior in ex-ante forecasting a crisis. Surprisingly however, it turned out that not all but those models which implemented features of the causes of the global financial crisis (like financial frictions or interbank credit flows) could not only detect the turning point of the Austrian business cycle early in 2008 but they also succeeded in forecasting the following severe recession in 2009. In comparison, non-DSGE methods like the ex-ante forecast with the Global Economic (Macro) Model of Oxford Economics and WIFO's expert forecasts performed not better than DSGE models in the crisis.
    Keywords: DSGE models, business cycles, forecasting, open-economy macroeconomics
    Date: 2016–11–08
  4. By: Gonçalo Faria (Católica Porto Business School and CEGE, Universidade Católica Portuguesa); Fabio Verona (Bank of Finland and CEF.UP)
    Abstract: We forecast stock market returns by applying, within a Ferreira and Santa-Clara (2011) sum-of-the-parts framework, a frequency decomposition of several predictors of stock returns. The method delivers statistically and economically significant improvements over historical mean forecasts, with monthly out- of-sample R2 of 3.27% and annual utility gains of 403 basis points. The strong performance of this method comes from its ability to isolate the frequencies of the predictors with the highest predictive power from the noisy parts, and from the fact that the frequency-decomposed predictors carry complementary information that captures both the long-term trend and the higher frequency movements of stock market returns.
    Keywords: predictability, stock returns, equity premium, asset allocation, frequency domain, wavelets
    JEL: G11 G12 G14 G17
    Date: 2016–10
  5. By: Martin, Ian; Wagner, Christian
    Abstract: We derive a formula that expresses the expected return on a stock in terms of the risk-neutral variance of the market and the stock's excess risk-neutral variance relative to the average stock. These components can be computed from index and stock option prices; the formula has no free parameters. We test the theory in-sample by running panel regressions of stock returns onto risk-neutral variances. The formula performs well at 6-month and 1-year forecasting horizons, and our predictors drive out beta, size, book-to-market, and momentum. Out-of-sample, we find that the formula outperforms a range of competitors in forecasting individual stock returns. Our results suggest that there is considerably more variation in expected returns, both over time and across stocks, than has previously been acknowledged.
    Keywords: expected returns; forecast; implied volatility; risk premia; risk-neutral variance
    JEL: E22 E44 G10 G12 G17 G31 G32
    Date: 2016–11
  6. By: Markus Dabernig; Georg J. Mayr; Jakob W. Messner; Achim Zeileis
    Abstract: Statistical post-processing of ensemble predictions is usually adjusted to a particular lead time so that several models must be fitted to forecast multiple lead times. To increase the coherence between lead times, we propose to use standardized anomalies instead of direct observations and predictions. By subtracting a climatological mean and dividing by the climatological standard deviation, lead-time-specific characteristics are eliminated and several lead times can be forecasted simultaneously. The results show that forecasts between +12 and +120 h can be fitted together with a comparable forecast skill to a conventional method. Furthermore, forecasts can be produced with a temporal resolution as high as the observation interval e.g., up to ten minutes.
    Keywords: standardized anomalies, non-homogeneous regression, ensemble post-processing, probabilistic temperature forecasts
    JEL: C53 C61 Q50
    Date: 2016–10

This nep-for issue is ©2016 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.