nep-for New Economics Papers
on Forecasting
Issue of 2012‒06‒13
thirteen papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting national recessions using state-level data By Owyang, Michael T.; Piger, Jeremy; Wall, Howard J.
  2. Housing price forecastability: A factor analysis By Lasse Bork; Stig V. Møller
  3. Empirical simultaneous prediction regions for path-forecasts By Òscar Jordá; Malte Knuppel; Massimiliano Marcellino
  4. Macroeconomic forecasting during the Great Recession: The return of non-linearity? By Ferrara, L.; Marcellino, M.; Mogliani, M.
  5. Oil and US GDP: A Real-Time out-of Sample Examination By Francesco Ravazzolo; Philip Rothman
  6. WP 14 Revisions in official data and forecasting By Cecilia Frale; Valentina Raponi
  7. Nowcasting GDP in Real-Time: A Density Combination Approach By Knut Are Aastveit; Karsten R. Gerdrup; Anne Sofie Jore; Leif Anders Thorsrud
  8. Industry Effects on Firm and Segment Profitability Forecasting: Do Aggregation and Diversity Matter? By Yim, Andrew; Schröder, David
  9. WALS Prediction By Magnus, J.R.; Wang, W.; Zhang, Xinyu
  10. Spin model with negative absolute temperatures for stock market forecasting By J. L. Subias
  11. Content analysis of XBRL filings as an efficient supplement of bankruptcy prediction? Empirical evidence based on US GAAP annual reports By Henselmann, Klaus; Scherr, Elisabeth
  12. Does Aggregate Riskiness Predict Future Economic Downturns? By Bali, Turan G.; Cakici, Nusret; Chabi-Yo, Fousseni
  13. Is the Price Right? Assessing Estimates of Cadastral Values for Bogotá, Colombia By Luc Anselin; Nancy Lozano-Gracia

  1. By: Owyang, Michael T.; Piger, Jeremy; Wall, Howard J.
    Abstract: A large literature studies the information contained in national-level economic indicators, such as nancial and aggregate economic activity variables, for forecasting U.S. business cycle phases (expansions and recessions.) In this paper, we investigate whether there is additional information regarding business cycle phases contained in subnational measures of economic activity. Using a probit model to predict the NBER expansion and recession classification, we assess the forecasting benets of adding state-level employment growth to a common list of national-level predictors. As state-level data adds a large number of variables to the model, we employ a Bayesian model averaging procedure to construct forecasts. Based on a variety of forecast evaluation metrics, we find that including state-level employment growth substantially improves short-horizon forecasts of the business cycle phase. The gains in forecast accuracy are concentrated during months of national recession. Posterior inclusion probabilities indicate substantial uncertainty regarding which states belong in the model, highlighting the importance of the Bayesian model averaging approach.
    Keywords: turning points; probit; covariate selection
    JEL: C53 E32 C52
    Date: 2012–04–10
  2. By: Lasse Bork (Aalborg University); Stig V. Møller (Aarhus University and CREATES)
    Abstract: We examine US housing price forecastability using a common factor approach based on a large panel of 122 economic time series. We find that a simple three-factor model generates an explanatory power of about 50% in one-quarter ahead in-sample forecasting regressions. The predictive power of the model stays high at longer horizons. The estimated factors are strongly statistically signi?cant according to a bootstrap resampling method which takes into account that the factors are estimated regressors. The simple three-factor model also contains substantial out-of-sample predictive power and performs remarkably well compared to both autoregressive benchmarks and computational intensive forecast combination models.
    Keywords: House prices, Forecasting, Factor model, Principal components, Macroeconomic factors, Factor forecast combination, Bootstrap
    JEL: C53 E3 G1
    Date: 2012–05–25
  3. By: Òscar Jordá; Malte Knuppel; Massimiliano Marcellino
    Abstract: This paper investigates the problem of constructing prediction regions for forecast trajectories 1 to H periods into the future - a path forecast. We take the more general view that the null model is only approximative and in some cases it may be altogether unavailable. As a consequence, one cannot derive the usual analytic expressions nor resample from the null model as is usually done when bootstrap methods are used. The paper derives methods to construct approximate rectangular regions for simultaneous probability coverage which correct for serial correlation. The techniques appear to work well in simulations and in an application to the Greenbook path-forecasts of growth and inflation.
    Keywords: Forecasting
    Date: 2012
  4. By: Ferrara, L.; Marcellino, M.; Mogliani, M.
    Abstract: The debate on the forecasting ability in economics of non-linear models has a long history, and the Great Recession provides us with an opportunity for a re-assessment of the forecasting performance of several classes of non-linear models, widely used in applied macroeconomic research. In this paper, we carry out an extensive analysis over a large quarterly database consisting of major real, nominal and financial variables for a large panel of OECD member countries. It turns out that, on average, non-linear models do not outperform standard linear specifications, even during the Great Recession period. In spite of this result, non-linear models enable to improve forecast accuracy in almost 40% of cases. Especially some countries and/or variables appear to be more adapted to non-linear forecasting.
    Keywords: Forecasting, Non-linear models, Great Recession.
    JEL: C22 C53 E37
    Date: 2012
  5. By: Francesco Ravazzolo; Philip Rothman
    Abstract: We study the real-time predictive content of crude oil prices for US real GDP growth through a pseudo out-of-sample (OOS) forecasting exercise. Comparing our benchmark model “withoutoil†against alternatives “with oil,†we strongly reject the null hypothesis of no OOS population-level predictability from oil prices to GDP at the longer forecast horizon we consider. These results may be due to our oil price measures serving as proxies for a recently developed measure of global real economic activity omitted from the alternatives to the benchmark forecasting models. This examination of the global OOS relative performance of the models we consider is robust to use of ex-post revised data. But when we focus on the forecasting models’ local relative performance, we observe strong differences across use of real-time and ex-post revised data.
    JEL: C22 C53 E32 E37
    Date: 2011–11
  6. By: Cecilia Frale; Valentina Raponi
    Abstract: This paper deals with the topic of revision of data with the aim of investigating whether consecutive releases of macroeconomic series published by statistical agencies contain useful information for economic analysis and forecasting. The rationality of the re-visions process is tested considering the complete history of data and an empirical application to show the usefulness of revisions for improving the precision of forecasting model is proposed. The results for Italian GDP growth show that embedding the revision process in a dynamic factor model helps to reduce the forecast error.
    Keywords: Data revisions, real-time dataset, mixed frequency, Dynamic factor Model
    JEL: E32 E37 C53
    Date: 2012–03
  7. By: Knut Are Aastveit; Karsten R. Gerdrup; Anne Sofie Jore; Leif Anders Thorsrud
    Abstract: In this paper we use U.S. real-time vintage data and produce combined density nowcasts for quarterly GDP growth from a system of three commonly used model classes. The density nowcasts are combined in two steps. First, a wide selection of individual models within each model class are combined separately. Then, the nowcasts from the three model classes are combined into a single predictive density. We update the density now-cast for every new data release throughout the quarter, and highlight the importance of new information for the evaluation period 1990Q2-2010Q3. Our results show that the logarithmic score of the predictive densities for U.S. GDP increase almost monotonically as new information arrives during the quarter. While the best performing model class is changing during the quarter, the density nowcasts from our combination framework is always performing well both in terms of logarithmic scores and calibration tests. The density combination approach is superior to a simple model selection strategy and also performs better in terms of point forecast evaluation than standard point forecast combinations.
    Keywords: Density combination; Forecast densities; Forecast evaluation; Monetary policy; Nowcasting; Real-time data
    JEL: C32 C52 C53 E37 E52
    Date: 2011–09
  8. By: Yim, Andrew; Schröder, David
    Abstract: Abstract. A recent study shows that industry-specific analysis has no incremental advantage over economy-wide analysis in forecasting firm profitability. This result seems puzzling because some earlier studies have documented the importance of industry effects in explaining firm profitability. We reconcile the apparent inconsistency by showing that industry effects on profitability forecasting exist at the more refined business segment level, but are obscured by aggregated reporting at the firm level. Using segment-level analysis as well as firm-level analysis that also utilizes segment-level information, we provide consistent evidence supporting that industry-specific analysis is more accurate than economy-wide analysis in predicting the profitability of business segments and the profitability of single-segment firms.
    Keywords: Segment profitability; Earnings predictability; Earnings persistence; Aggregation; Diversity; Industry membership
    JEL: L25 M41 G00 M21
    Date: 2012–06–01
  9. By: Magnus, J.R.; Wang, W.; Zhang, Xinyu (Tilburg University, Center for Economic Research)
    Abstract: Abstract: Prediction under model uncertainty is an important and difficult issue. Traditional prediction methods (such as pretesting) are based on model selection followed by prediction in the selected model, but the reported prediction and the reported prediction variance ignore the uncertainty from the selection procedure. This paper proposes a weighted average least squares (WALS) prediction procedure that is not conditional on the selected model. Taking both model and error uncertainty into account, we also propose an appropriate estimate of the variance of the WALS predictor. Correlations among the random errors are explicitly allowed. Compared to other prediction averaging methods, the WALS predictor has important advantages both theoretically and computationally. Simulation studies show that the WALS predictor generally produces lower mean squared prediction errors than its competitors, and that the proposed estimator for the prediction variance performs particularly well when model uncertainty increases.
    Keywords: Model averaging;Model uncertainty;Bayesian analysis;Prediction.
    Date: 2012
  10. By: J. L. Subias
    Abstract: A spin model relating physical to financial variables is presented. Based on this model, an algorithm evaluating negative temperatures was applied to New York Stock Exchange quotations from May 2005 up to the present. Stylized patterns resembling known processes in phenomenological thermodynamics were found, namely, population inversion and the magnetocaloric effect.
    Date: 2012–06
  11. By: Henselmann, Klaus; Scherr, Elisabeth
    Abstract: Most of the bankruptcy prediction models developed so far have in common that they are based on quantitative data or more precisely financial ratios. However, useful information can be lost when disregarding soft information. In this work, we develop an automated content analysis technique to assess the bankruptcy risk of companies using XBRL tags. We develop a list of potential red flags based on the U.S. GAAP taxonomy and assign the elements to 2 categories and 7 subcategories. Then we test our red flag item list based on U.S. GAAP annual reports of 26 companies with Chapter 11 bankruptcy filings and a control group. The empirical results show that in total, the red flag item list has predictive power of bankruptcy risk. Logistic regression results also show that the predictive power increases the nearer the bankruptcy filing date approaches. We furthermore observe that the category 2 red flags (bankruptcy characteristics and influencing factors) have higher discriminatory power than category 1 red flags (earnings management indicators) for one year before the bankruptcy filing date. This difference narrows for two years before the bankruptcy filing date and may turn in favor of category 1 red flags for three years before the bankruptcy filing date. --
    Keywords: content analysis,red flags,XBRL,bankruptcy prediction,risk assessment,earnings management,Inhaltsanalyse,Red Flags,XBRL,Insolvenzprognose,Risikobewertung,Bilanzpolitik
    JEL: M41 C12 C81
    Date: 2012
  12. By: Bali, Turan G. (Georgetown University); Cakici, Nusret (Fordham University); Chabi-Yo, Fousseni (OH State University)
    Abstract: Aumann and Serrano (2008) and Foster and Hart (2009) introduce riskiness measures based on the physical return distribution of gambles. This paper proposes model-free options' implied measures of riskiness based on the risk-neutral distribution of financial securities. In addition to introducing the forward-looking measures of riskiness, the paper investigates the significance of aggregate riskiness in predicting future economic downturns. The results indicate strong predictive power of aggregate riskiness even after controlling for the realized volatility of the U.S. equity market, the implied volatility of S&P 500 index options (VIX) proxying for financial market uncertainty, as well as the TED spread proxying for interbank credit risk and the perceived health of the banking system.
    JEL: G11 G12 G14 G33
    Date: 2012–05
  13. By: Luc Anselin (GeoDa Center for Geospatial Analysis and Computation; Arizona State University); Nancy Lozano-Gracia (GeoDa Center for Geospatial Analysis and Computation; Arizona State University)
    Abstract: Hedonic house price models are increasingly applied in the process of mass appraisal, in which econometric specifications are used to obtain automated valuation of properties for taxation purposes. The predictive quality of such models is important, since it directly affects the revenue stream of local authorities. In this paper, we assess the relative predictive performance of different model specifications used in automated valuation. Specifically, we focus on the issue of spatial heterogeneity by comparing models that utilize different definitions of housing submarkets. In addition, we consider the inclusion of “spatial†explanatory variables in the form of distance to various amenities as computed from a GIS. We apply this to data from the city of Bogot Ìa, Colombia, a pioneer in the application of mass appraisal techniques in a developing country context. We find that specifications that include the submarkets improve predictive performance and that the inclusion of the spatial variables is superior to the traditional models of homogenous zones. However, even the best models are still characterized by relatively poor performance in the form of a high degree of overprediction of the house value. In addition, the predictive performance of the models varied by socio-economic stratum in the city, which suggests that the dynamics of the housing markets in these strata would require closer and separate attention. These results may provide further guidance to enhance mass appraisal practice in the city of Bogot Ìa as well as potentially other Latin American cities.
    Date: 2011

This nep-for issue is ©2012 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.