nep-for New Economics Papers
on All new papers
Issue of 2014‒09‒08
thirteen papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting US Real Private Residential Fixed Investment Using a Large Number of Predictors By Goodness C. Aye; Stephen M. Miller; Rangan Gupta; Mehmet Balcilar
  2. Forecasting the U.S. Real House Price Index By Vasilios Plakandaras; Rangan Gupta; Periklis Gogas; Theophilos Papadimitriou
  3. Forecasting South African Ination Using Non-linear Models: A Weighted Loss-based Evaluation By Pejman Bahramian; Mehmet Balcilar; Rangan Gupta; Patrick T. Kanda
  4. Forecasting the Price of Gold By Hossein Hassani; Emmanuel Sirimal Silva; Rangan Gupta
  5. Combining distributions of real-time forecasts: An application to U.S. growth By Götz T.B.; Hecq A.W.; Urbain J.R.Y.J.
  6. Forecasting the Price of Gold Using Dynamic Model Averaging By Goodness Aye; Rangan Gupta; Shawkat Hammoudeh; Won Joong Kim
  7. DSGE Model-Based Forecasting of Modeled and Non-Modeled Ination Variables in South Africa By Rangan Gupta; Patrick T. Kanda; Mampho P. Modise; Alessia Paccagnini
  8. Forecasting US Real House Price Returns over 1831-2013: Evidence from Copula Models By Anandamayee Majumdar; Rangan Gupta
  9. Nonparametric Estimation of Conditional Expectations for Sustainability Analyses By Alho, Juha
  10. True or Spurious Long Memory in Volatility : Further Evidence on the Energy Futures Markets By Charfeddine Lanouar
  11. Eliciting and aggregating individual expectations: An experimental study By Peeters R.J.A.P.; Wolk K.L.
  12. Testing for Granger causality in large mixed-frequency VARs By Götz T.B.; Hecq A.W.
  13. Has Oil Pirce Predicted Stock Returns for Over a Century? By Paresh K. Narayan; Nico Katzke

  1. By: Goodness C. Aye; Stephen M. Miller; Rangan Gupta; Mehmet Balcilar
    Abstract: This paper employs classical bivariate, factor augmented (FA), slab-and-spike variable selection (SSVS)-based, and Bayesian semi-parametric shrinkage (BSS)-based predictive regression models to forecast US real private residential fixed investment over an out-ofsample period from 1983:Q1 to 2011:Q2, based on an in-sample estimates for 1963:Q1 to 1982:Q4. Both large-scale (188 macroeconomic series) and small-scale (20 macroeconomic series) FA, SSVS, and BSS predictive regressions, as well as 20 bivariate regression models, capture the influence of fundamentals in forecasting residential investment. We evaluate the ex-post out-of-sample forecast performance of the 26 models using the relative average Mean Square Error for one-, two-, four-, and eight-quarters-ahead forecasts and test their significance based on the McCracken (2004, 2007) MSE-F statistic. We find that, on average, the SSVS-Large model provides the best forecasts amongst all the models. We also find that one of the individual regression models, using house for sale (H4SALE) as a predictor, performs best at the four- and eight-quarters-ahead horizons. Finally, we use these two models to predict the relevant turning points of the residential investment, via an ex-ante forecast exercise from 2011:Q3 to 2012:Q4. The SSVS-Large model forecasts the turning points more accurately, although the H4SALE model does better toward the end of the sample. Our results suggest that economy-wide factors, in addition to specific housing market variables, prove important when forecasting in the real estate market.
    Keywords: Private residential investment, predictive regressions, factoraugmented models, Bayesian shrinkage, forecasting
    JEL: C32 E22 E27
    Date: 2014–08–29
  2. By: Vasilios Plakandaras; Rangan Gupta; Periklis Gogas; Theophilos Papadimitriou
    Abstract: The 2006 sudden and immense downturn in U.S. House Prices sparked the 2007 global financial crisis and revived the interest about forecasting such imminent threats for economic stability. In this paper we propose a novel hybrid forecasting methodology that combines the Ensemble Empirical Mode Decomposition (EEMD) from the field of signal processing with the Support Vector Regression (SVR) methodology that originates from machine learning. We test the forecasting ability of the proposed model against a Random Walk (RW) model, a Bayesian Autoregressive and a Bayesian Vector Autoregressive model. The proposed methodology outperforms all the competing models with half the error of the RW model with and without drift in out-of-sample forecasting. Finally, we argue that this new methodology can be used as an early warning system for forecasting sudden house prices drops with direct policy implications.
    Keywords: house prices, forecasting, machine learning, Support Vector Regression.
    JEL: C32 C53 R31
    Date: 2014–08–29
  3. By: Pejman Bahramian; Mehmet Balcilar; Rangan Gupta; Patrick T. Kanda
    Abstract: The conduct of in ation targeting is heavily dependent on accurate in ation forecasts. Non-linear models have increasingly featured, along with linear counterparts, in the forecasting literature. In this study, we focus on forecasting South African in ation by means of non-linear models and using a long historical dataset of seasonally-adjusted monthly in ation rates spanning from 1921:02 to 2013:01. For an emerging market economy such as South Africa, non-linearities can be a salient feature of such long data, hence the relevance of evaluating non-linear models' forecast performance. In the same vein, given the fact that 1969:10 marks the beginning of a protracted rising trend in South African in ation data, we estimate the models for an in-sample period of 1921:02-1966:09 and evaluate 24 step-ahead forecasts over an out-of-sample period of 1966:10-2013:01. In addition, using a weighted loss function specication, we evaluate the forecast performance of dierent non-linear models across various extreme economic environments and forecast horizons. In general, we nd that no competing model consistently and signicantly beats the LoLiMoT's performance in forecasting South African in ation.
    Keywords: In ation, forecasting, non-linear models, weighted loss function, South Africa
    JEL: C32 E31 E52
    Date: 2014–08–29
  4. By: Hossein Hassani; Emmanuel Sirimal Silva; Rangan Gupta
    Abstract: This paper seeks to evaluate the appropriateness of a variety of existing forecasting techniques (17 methods) at providing accurate, and statistically significant forecasts for gold price. We report the results from the 9 most competitive techniques. Special consideration is given to the ability of these techniques at providing forecasts which outperforms the random walk as we noticed that certain multivariate models (which included prices of silver, platinum, palladium and rhodium, besides gold) were also unable to outperform the random walk in this case. Interestingly, the results show that none of the forecasting techniques are able to outperform the random walk at horizons of 1 and 9 steps ahead, and on average the Exponential Smoothing model is seen providing the best forecasts in terms of the lowest root mean squared error over the 24 months forecasting horizons. Moreover, we find that the univariate models used in this paper are able to outperform the Bayesian autoregression, and Bayesian vector autoregressive models, with exponential smoothing (ETS) reporting statistically significant results in comparison to the former models, and classical autoregressive and the vector autoregressive models in most cases.
    Keywords: ARIMA; ETS; TBATS; ARFIMA; AR; VAR; BAR; BVAR; Random Walk; Gold; Forecast; Multivariate; Univariate.
    Date: 2014–08–29
  5. By: Götz T.B.; Hecq A.W.; Urbain J.R.Y.J. (GSBE)
    Abstract: We extend the repeated observations forecasting ROF analysis of Croushore and Stark 2002 to allow for regressors of possibly higher sampling frequencies than the regressand. For the U.S. GNP quarterly growth rate, we compare the forecasting performances of an AR model with several mixed-frequency models among which is the MIDAS approach. Using the additional dimension provided by different vintages we compute several forecasts for a given calendar date and subsequently approximate the corresponding distribution of forecasts by a continuous density. Scoring rules are then employed to construct combinations of them and analyze the composition and evolvement of the implied weights over time. Using this approach, we not only investigate the sensitivity of model selection to the choice of which data release to consider, but also illustrate how to incorporate revision process information into real-time studies. As a consequence of these analyses, weintroduce a new weighting scheme that summarizes information contained in the revision process of the variables under consideration.
    Keywords: Single Equation Models; Single Variables: Models with Panel Data; Longitudinal Data; Spatial Time Series; Forecasting and Prediction Methods; Simulation Methods ;
    JEL: C23 C53
    Date: 2014
  6. By: Goodness Aye; Rangan Gupta; Shawkat Hammoudeh; Won Joong Kim
    Abstract: We develop models for examining possible predictors of the return on gold that embrace six global factors (business cycle, nominal, interest rate, commodity, exchange rate and stock price factors) and two uncertainty indices (the Kansas City Fed’s financial stress index and the U.S. Economic uncertainty index). Specifically, by comparing with other alternative models, we show that the dynamic model averaging (DMA) and dynamic model selection (DMS) models outperform not only a linear model (such as random walk) but also the Bayesian model averaging (BMA) model for examining possible predictors of the return of gold. The DMS is the best overall across all forecast horizons. Generally, all the predictors show strong predictive power at one time or another though at varying magnitudes, while the exchange rate factor and the Kansas City Fed’s financial stress index appear to be strong at almost all horizons and sub-periods. However, the forecasting prowess of the exchange rate is supreme.
    Keywords: Bayesian, state space models, gold, macroeconomic fundamentals, forecasting
    JEL: C11 C53 F37 F47 Q02
    Date: 2014–08–29
  7. By: Rangan Gupta; Patrick T. Kanda; Mampho P. Modise; Alessia Paccagnini
    Abstract: In ation forecasts are a key ingredient for monetary policymaking - especially in an in ation targeting country such as South Africa. Generally, a typical Dynamic Stochastic General Equilibrium (DSGE) only includes a core set of variables. As such, other variables,e.g. such as alternative measures of in ation that might be of interest to policymakers, do not feature in the model. Given this, we implement a closed-economy New Keynesian DSGE model-based procedure which includes variables that do not explicitly appear in the model. We estimate such a model using an in-sample covering 1971Q2 to 1999Q4, and generate recursive forecasts over 2000Q1-2011Q4. The hybrid DSGE performs extremely well in forecasting in ation variables (both core and non-modeled) in comparison with forecasts reported by other models such as AR(1).
    Keywords: DSGE model, in ation, core variables, non-core variables
    JEL: C11 C32 C53 E27 E47
    Date: 2014–08–29
  8. By: Anandamayee Majumdar (Center of Advanced Statistics and Econometrics, Soochow University, China); Rangan Gupta (Department of Economics, University of Pretoria)
    Abstract: Given the existence of non-normality and nonlinearity in the data generating process of real house price returns over the period of 1831-2013, this paper compares the ability of various univariate copula models, relative to standard benchmarks (naive and autoregressive models) in forecasting real US house price over the annual out-of-sample period of 1859-2013, based on an in-sample of 1831-1858. Overall, our results provide overwhelming evidence in favor of the copula models (Normal, Student’s t, Clayton, Frank, Gumbel, Joe and Ali-Mikhail-Huq) relative to linear benchmarks, and especially for the Student’s t copula, which outperforms all other models both in terms of in-sample and out-of-sample predictability results. Our results highlight the importance of accounting for non-normality and nonlinearity in the data generating process of real house price returns for the US economy for nearly two centuries of data.
    Keywords: House Price, Copula Models, Forecasting
    JEL: C22 C53 R3
    Date: 2014–08
  9. By: Alho, Juha
    Abstract: Optimal forecasts are, under a squared error loss, conditional expectations of the unknown future values of interest. When stochastic demographic models are used in macroeconomic analyses, it becomes important to be able to handle updated forecasts. That is, when population development turns out to differ from the expected one, the decision makers in the macroeconomic models may change their behavior. To allow for this, numerical methods have been developed that allow us to approximate how future forecasts might look like, for any given observed path. Some technical details of how this can be done in the R environment are given.
    Keywords: demography, forecasting, overlapping generations
    Date: 2014–08–25
  10. By: Charfeddine Lanouar
    Abstract: The main goal of this paper is to investigate whether the long memory behavior observed in many volatility energy futures markets series is a spurious behavior or not. For this purpose, we employ a wide variety of advanced volatility models that allow for long memory and/or structural changes : the GARCH(1,1), the FIGARCH(1,d,1), the Adaptative-GARCH(1,1,k), and the Adaptative-FIGARCH(1,d,1,k) models. To compare forecasting ability of these models, we use out-of- sample forecasting performance. Using the crude oil, heating oil, gaso- line and propane volatility futures energy time series with one month and three month's maturities, we found that ve out of the eight time series are characterized by both long memory and structural breaks. For these series, dates of breaks coincide with some majors economics and nancial events. For the three others time series, we found strong evidence of long memory in volatility.
    Keywords: Long Memory, Structural Breaks, Fractional Integra- tion, Volatility, Volatility Forecasting.
    Date: 2014–08–29
  11. By: Peeters R.J.A.P.; Wolk K.L. (GSBE)
    Abstract: In this paper we present a mechanism to elicit and aggregate dispersed information. Our mechanism relies on the aggregation of intervals elicited using an interval scoring rule. We test our mechanism by eliciting beliefs about the termination times of a stochastic process in an experimental setting. We conduct two treatments, one with high and one with low volatility.Increasing the underlying volatility affects the location of the interval, yet it does not significantly affect its length. Consequently, individuals perform significantly better in the low volatility treatment than in the high volatility treatment. Next, we construct distributions by aggregating intervals across different individuals. Our results reveal that the predictive quality of the aggregated intervals as measured by the Hellinger distance to the true distribution increases by more than 30 when increasing the aggregation level from two to eight individuals. This shows that aggregating individual intervals may be an attractive solution when market mechanisms are infeasible.
    Keywords: Forecasting and Prediction Methods; Simulation Methods ; Design of Experiments: Laboratory, Individual; Expectations; Speculations;
    JEL: C53 D84 C91
    Date: 2014
  12. By: Götz T.B.; Hecq A.W. (GSBE)
    Abstract: In this paper we analyze Granger causality testing in a mixed-frequency VAR, originally proposed by Ghysels 2012, where the difference in sampling frequencies of the variables is large. In particular, we investigate whether past information on a low-frequency variable help in forecasting a high-frequency one and vice versa. Given a realistic sample size, the number of high-frequency observations per low-frequency period leads to parameter proliferation problems in case we attempt to estimate the model unrestrictedly. We propose two approaches to solve this problem, reduced rank restrictions and a Bayesian mixed-frequency VAR. For the latter, we extend the approach in Banbura et al. 2010 to a mixed-frequency setup, which presents an alternative to classical Bayesian estimation techniques. We compare these methods to a common aggregated low-frequency model as well as to the unrestricted VAR in terms of their Granger non-causality testing behavior using Monte Carlo simulations. The techniques are illustrated in an empirical application involving dailyrealized volatility and monthly business cycle fluctuations.
    Keywords: Hypothesis Testing: General; Multiple or Simultaneous Equation Models: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models;
    JEL: C12 C32
    Date: 2014
  13. By: Paresh K. Narayan (Centre for Financial Econometrics, School of Accounting, Economics and Finance, Deakin University, Australia); Nico Katzke (Department of Economics, University of Pretoria)
    Abstract: This paper contributes to the debate on the role of oil prices in predicting stock returns. The novelty of the paper is that it considers monthly time-series historical data that span over 150 years (1859:10-2013:12) and applies a predictive regression model that accommodates three salient features of the data, namely, a persistent and endogenous oil price, and model heteroskedasticity. Three key findings are unraveled: First, oil price predicts US stock returns. Second, in-sample evidence is corroborated by out-sample evidence of predictability. Third, both positive and negative oil price changes are important predictors of US stock returns, with negative changes relatively more important. Our results are robust to the use of different estimators and choice of in-sample periods.
    Keywords: Stock returns, Predictability, Oil price
    JEL: C22 E37 G17 Q43
    Date: 2014–08

This nep-for issue is ©2014 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.