nep-for New Economics Papers
on Forecasting
Issue of 2014‒09‒05
eleven papers chosen by
Rob J Hyndman
Monash University

  1. Nowcasting and Forecasting Economic Growth in the Euro Area using Principal Components By Irma Hindrayanto; Siem Jan Koopman; Jasper de Winter
  2. Forecasting Exchange Rates under Model and Parameter Uncertainty By Joscha Beckmann; Rainer Schüssler
  3. The FRBNY Staff Underlying Inflation Gauge: UIG By Marlene Amstad; Simon Potter; Robert Rich
  4. Forecasting with DSGE models with financial frictions By Michał Rubaszek; Marcin Kolasa
  5. The Costs of Error in Setting Reference Rates for Reduced Deforestation By Patrick Doupe
  6. Short and long-term forecasting by the Netherlands Bureau for Economic Policy Analysis (CPB): science, witchcraft, or practical tool for policy? By Bos, Frits; Teulings, Coen
  7. An estimated DSGE model of a Small Open Economy within the Monetary Union: Forecasting and Structural Analysis By Yuliya Rychalovska; Massimiliano Marcellino (EUI)
  8. Density Forecasting using Bayesian Global Vector Autoregressions with Common Stochastic Volatility By Florian Huber
  9. Going granular: The importance of firm-level equity information in anticipating economic activity By Filippo di Mauro; Filippo di Mauro, Fabio Fornari
  10. Forecasting the sales of an innovative agro-industrial product with limited information: A case of feta cheese from buffalo milk in Thailand By Komsan Suriya; Orakanya Kanjanatarakul
  11. Can Google searches help nowcast and forecast unemployment rates in the Visegrad Group countries? By Jaroslav Pavlicek; Ladislav Kristoufek

  1. By: Irma Hindrayanto (De Nederlandsche Bank); Siem Jan Koopman (VU University Amsterdam, the Netherlands); Jasper de Winter (De Nederlandsche Bank, the Netherlands)
    Abstract: Many empirical studies have shown that factor models produce relatively accurate forecasts compared to alternative short-term forecasting models. These empirical findings have been established for different macroeconomic data sets and different forecast horizons. However, various specifications of the factor model exist and it is a topic of debate which specification is most effective in its forecasting performance. Furthermore, the forecast performances of the different specifications during the recent financial crisis are also not well documented. In this study we investigate these two issues in depth. We empirically verify the forecast performance of three factor model approaches and report our findings in an extended empirical out-of-sample forecasting competition for quarterly growth of gross domestic product in the euro area and its five largest countries over the period 1992-2012. We also introduce two extensions of existing factor models to make them more suitable for real-time forecasting. We show that the factor models have been able to systematically beat the benchmark autoregressive model, both before as well as during the financial crisis. The recently proposed collapsed dynamic factor model shows the highest forecast accuracy for the euro area and the majority of countries that we have analyzed. The forecast precision improvements against the benchmark model can range up to 77% in mean square error reduction, depending on the country and forecast horizon.
    Keywords: Factor models; Principal component analysis; Forecasting; Kalman filter; State space method; Publication lag; Mixed frequency
    JEL: C32 C53 E17
    Date: 2014–08–22
  2. By: Joscha Beckmann; Rainer Schüssler
    Abstract: We introduce a forecasting method that closely matches the econometric properties required by the theory on exchange rate prediction. Our approach formally models (i) when (and if) explanatory variables enter or leave a regression model, (ii) the degree of parameter instability, (iii) the (potentially) rapidly changing relevance of regressors, and (iv) the appropriate shrinkage intensity over time. We consider (short-term) forecasting of six major US dollar exchange rates using a standard set of macro fundamentals. Our results indicate the importance of shrinkage and flexible model selection/averaging criteria to avoid poor forecasting results.
    Keywords: Exchange rates forecasting, time-varying parameter models, shrinkage, model selection/averaging
    JEL: F31 F37 G17
    Date: 2014–08
  3. By: Marlene Amstad; Simon Potter; Robert Rich
    Abstract: Monetary policymakers and long-term investors would benefit greatly from a measure of underlying inflation that uses all relevant information, is available in real-time, and forecasts inflation better than traditional underlying inflation measures such as core inflation measures. This paper presents the "Federal Reserve Bank of New York (FRBNY) Staff Underlying Inflation Gauge (UIG)" for CPI and PCE. Using a dynamic factor model approach, the UIG is derived from a broad data set that extends beyond price series to include a wide range of nominal, real, and financial variables. It also considers the specific and time-varying persistence of individual subcomponents of an inflation series. An attractive feature of the UIG is that it can be updated on a daily basis, which allows for a close monitoring of changes in underlying inflation. This capability can be very useful when large and sudden economic fluctuations occur, as at the end of 2008. In addition, the UIG displays greater forecast accuracy than traditional measures of core inflation.
    Keywords: Inflation, Dynamic Factor Models, Core Inflation, Monetary Policy, Forecasting
    Date: 2014–07
  4. By: Michał Rubaszek; Marcin Kolasa
    Abstract: To investigate to what extent adding financial frictions can contribute to an improvement in the quality of DSGE model-based forecasts DSGE models with and without financial frictions. Comparison of point and density forecasts. The main finding is that accounting for financial frictions affecting firms tends to improve the quality of point forecasts while the opposite is true for the extension with household sector financial frictions. However, for all models point forecasts can be considered poor in the absolute sense and density forecasts are rather badly calibrated. We show that the main source of these problems is a significant and sizable bias in the forecasts for most of standard macroeconomic variables.
    Keywords: United States, General equilibrium modeling, Forecasting and projection methods
    Date: 2013–06–21
  5. By: Patrick Doupe (Potsdam Institute for Climate Impact Research)
    Abstract: To measure the deforestation reduced by a policy, we need to compare deforestation rates under a policy with deforestation rates in the absence of policy. Unfortunately the deforestation rate in the absence of a policy, or reference rate, is ex ante difficult to forecast and ex post impossible to observe. This means that reference rates will be set with error and we will not know how large the error will be. The challenging nature of setting reference rates is reflected in the number of proposals for reference rate design. In this paper I show how these proposals ignore forecast error. As a consequence, these proposals have basic structural weaknesses that in- crease the costs of reduced deforestation policy. I propose that a criteria for reference rates is to minimise the cost of forecast error. These ideas are illustrated with a cross country dataset on agricultural expansion. I show that the best forecasting model differs by country and that a countryÕs best forecasting model can be very simple.
    Keywords: climate policy, reduced deforestation, forecasting
    JEL: C53 Q5 Q57
    Date: 2014–08
  6. By: Bos, Frits; Teulings, Coen
    Abstract: This paper discusses five different types of forecasts by CPB: forecasts for next year, forecasts for next period of government, analyses of the sustainability of public finance, long-term scenarios and long-term effects of election platforms. CPB forecasts for next year and for the next period of government should be seen as well-motivated estimates based on all recent information, plausible assumptions and expected trends. The more distant the look into the future, the more uncertain are the forecasts. For such long-term analyses, the CPB employs scenarios, extended sensitivity analyses and identification of major political choices. Policy making is like sailing in fog. The regular set of CPB forecasts helps to look forward and to monitor whether a change of course is necessary. Despite fundamental uncertainty about the future, the CPB forecasts provide a good base for political discussions and decision making, like a coalition agreement, budget and wage rate negotiations and defining a long-term policy strategy. These forecasts inform Dutch society, reduce transaction costs in economic and political decision making, and foster consensus on economic and fiscal policy.
    Keywords: macroeconomic forecasting and government policy, accuracy of forecasts, uncertainty and public decision making, measurement in economics, CPB Netherlands Bureau for Economic Policy Analysis
    JEL: A11 C0 D8 E17 F17 F47 G17 H68
    Date: 2013
  7. By: Yuliya Rychalovska; Massimiliano Marcellino (EUI)
    Abstract: In this paper we build and estimate a two-region DSGE model of a small open economy within the European Monetary Union. We evaluate the properties of the estimated model and assess its forecasting performance (point and density) relative to reduced form models such as VARs. In addition, we study the empirical validity of the DSGE model restrictions (based on micro-foundations) by applying a DSGE-VAR approach. Finally, the estimated model is used to analyze the sources of macroeconomic fluctuations and examine the We allow for a sufficiently rich specification which enables us to include unemployment as well as open economy variables such as the real exchange rate into the estimation procedure, along with the standard macroeconomic and labor market indicators. The model contains a set of frictions and structural shocks typically used in the DSGE literature. We evaluate the properties of the estimated model and assess its forecasting performance relative to reduced form models such as VARs. In addition, we study the empirical validity of the DSGE model restrictions by applying a DSGE-VAR approach. Finally, the estimated model is used to analyze the sources of macroeconomic fluctuations and examine the responses of the economy to structural shocks. The model is built in the New Keynesian tradition and contains real and nominal rigidities such as habit formation in consumption, price and wage stickiness as well as rich stochastic structure. The framework also incorporates the theory of unemployment as in Gali et al. (2011), small open economy aspects and a nominal interest rate that is set exogenously by the area-wide monetary authority. The model is estimated using Bayesian techniques. We demonstrate that the estimated DSGE model is relatively well identified, has good data fit and reasonably estimated parameters. In addition, the model shows a competitive forecasting performance (in terms of both point and density) compared to reduced form models such as VARs. In this respect, our results are in line with the conclusions reached in previous studies that the new generation of DSGE models no longer faces the tension between rigor and fit. In particular, we illustrate that the DSGE model produces sizable one-step-ahead forecasting gains in terms of RMSE and the Score over the unrestricted VAR, especially for such variables as GDP, real exchange rate, unemployment and real wages. The predictions stay competitive at longer forecasting horizons. DSGE-VAR analysis demonstrates that the optimal weight on the DSGE restrictions is significant and the VAR(2) correction is not helpful in improving the DSGE model fit. At the same time, the DSGE-based prior significantly improves the short term forecast accuracy of the unrestricted VAR for output, and also determines a superior performance of the DSGE-VAR model in predicting exchange rate, unemployment and wages over all the forecast horizons considered here. When compared to an atheoretical Minnesota-style prior, the DSGE restrictions appear to be more useful in forecasting output and REER, whereas the opposite is true for employment. Application of the model to the analysis of the business cycle fluctuations demonstrates that "open economy" disturbances such as relative price, foreign demand and interest rate shocks explain a significant portion of the variation of output growth, inflation, real exchange rate and employment. Price and wage markup shocks are important determinants of inflation and real wages dynamics respectively.
    Keywords: Euro Area, Luxembourg, General equilibrium modeling, Forecasting and projection methods
    Date: 2013–06–21
  8. By: Florian Huber (Department of Economics, Vienna University of Economics and Business)
    Abstract: This paper puts forward a Bayesian Global Vector Autoregressive Model with Common Stochastic Volatility (B-GVAR-CSV). We assume that country specific volatility is driven by a single latent stochastic process, which simplifies the analysis and implies significant computational gains. Apart from computational advantages, this is also justified on the ground that the volatility of most macroeconomic quantities considered in our application tends to follow a similar pattern. Furthermore, Minnesota priors are used to introduce shrinkage to cure the curse of dimensionality. Finally, this model is then used to produce predictive densities for a set of macroeconomic aggregates. The dataset employed consists of quarterly data spanning from 1995:Q1 to 2012:Q4 and includes 45 economies plus the Euro Area. Our results indicate that stochastic volatility specifications influences accuracy along two dimensions: First, it helps to increase the overall predictive fit of our model. This result can be seen for some variables under scrutiny, most notably for real GDP and short-term interest rates. Second, it helps to make the model more resilient with respect to outliers and economic crises. This implies that when evaluated over time, the log predictive scores tend to show significantly less variation as compared to homoscedastic models.
    Keywords: Density Forecasting, Stochastic Volatility, Global vector autoregressions
    JEL: C32 F44 E32 E47
    Date: 2014–07
  9. By: Filippo di Mauro; Filippo di Mauro, Fabio Fornari
    Abstract: The paper attempts to verify whether equity returns of individual firms, and their realized volatilities, improve the in-sample and out-of-sample predictability of the US business cycle, as measured by the IP index VAR analysis and tests for forecasting ability The equity returns of individual firms, and their realized volatilities, are shown to improve the in-sample and out-of-sample predictability of the US business cycle, as measured by the IP index. In fact, significant declines in the root mean squared errors (RMSEs) are found when these variables are added to aggregate financial variables and selected macroeconomic indicators. Overall, to the aim of forecasting, there is a noticeable swing in the relative importance of individual firms across time, although firms that become key predictors of economic activity in a given month continue to do so for around six months, on average, bringing support to the idea that there is structure in the information that they convey. Unconditionally, belonging to a given sector does not boost the predictive power of firms, but we find that it becomes important for example around periods of recessions. Balance sheet data show that predictive ability of the firms is associated with features as performance, liquidity, the size of the foreign activity. Firm size also matters, as suggested by recent literature (Gabaix, 2011), although it is not - as put forward there - the only indicator to prevail.
    Keywords: European countries, Forecasting and projection methods, Microsimulation models
    Date: 2014–07–03
  10. By: Komsan Suriya; Orakanya Kanjanatarakul
    Abstract: This research forecasts the sales of an innovative agro-industrial product, the feta cheese from buffalo milk, in Thailand using limited information from January 2000 to August 2012. It aims to find how much data sufficiently needed for the prediction of accurate sales concerning that newly launched products are likely to provide insufficient information for traditional statistical methods. Furthermore, it compares two forecasting models; the Bass model (Bass, 1969) and the Logistic function (Stoneman, 2010) in terms of Mean Absolute Percentage Error (MAPE). It estimates the models by maximum likelihood and least squares methods with quadratic interpolation and quasi-Newton in Matlab. The findings show that the sales can be accurately forecasted by using monthly information just only 7 months to 24 months after the launching of the product. A comparison of the Bass model and the Logistic function using the same estimation methods shows that the Logistic function is superior to Bass model when using the data in the range of 7 to 24 months. In addition, the study indicates that the predictive sales values of the Bass model are always lower than those of the Logistic function. When combining them together, the Bass model always predicts the Lower-bound of the sales whereas those of Logistic function predicts the upper-bound. The area between the upper and lower bounds constructs the possibility of the sales. Last, the study calculates how long the product will last in the market and predicts the maximum sales by intercepting the lines of both models with the OLS linear time trend.
    Keywords: Thailand, Forecasting and projection methods, Modeling: new developments
    Date: 2013–06–21
  11. By: Jaroslav Pavlicek; Ladislav Kristoufek
    Abstract: Online activity of the Internet users has been repeatedly shown to provide a rich information set for various research fields. We focus on the job-related searches on Google and their possible usefulness in the region of the Visegrad Group -- the Czech Republic, Hungary, Poland and Slovakia. Even for rather small economies, the online searches of their inhabitants can be successfully utilized for macroeconomic predictions. Specifically, we study the unemployment rates and their interconnection to the job-related searches. We show that the Google searches strongly enhance both nowcasting and forecasting models of the unemployment rates.
    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.