|
on Forecasting |
| By: | Dovern, Jonas; Jannsen, Nils |
| Abstract: | Using real-time data, we analyze how the systematic expectation errors of professional forecasters in 19 advanced economies depend on the state of the business cycle. Our results indicate that the general result that forecasters systematically overestimate output growth (across all countries) masks considerable differences across different business-cycle states. We show that forecasts for recessions are subject to a large negative systematic forecast error (forecasters overestimate growth), while forecasts for recoveries are subject to a positive systematic forecast error. Forecasts made for expansions have, if anything, a small systematic forecast error for large forecast horizons. When we link information about the business-cycle state in the target year with quarterly information about its state in the forecasting period, we find that forecasters realize business-cycle turning points somewhat late. Using cross-country evidence, we demonstrate that the positive relationship between a change in trend growth rates and forecast bias, as suggested in the literature, breaks down when only focusing on forecasts made for expansions. |
| Keywords: | macroeconomic expectations,forecasting,forecast bias,survey data |
| JEL: | C5 E2 E3 |
| Date: | 2015 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:ifwkwp:1989&r=for |
| By: | Carlos Medel |
| Abstract: | Several years ago, the entire world experienced how fast and damaging certain inflationary shocks can be transmitted across seemingly uncorrelated countries. Despite the analysis of fuzzy transmission mechanisms, a direct inflationary transmission channel through global commodity prices shocks has been always of interest to policymakers—especially those concerned on imported inflation. The majority of international-to-domestic pass-through price measures are obviously insample estimations. However, in this article I analyse to what extent either global inflation or the Brent oil price provides more valuable information for future domestic inflation rates. I compare ten different multihorizon forecasts coming from a family of univariate time-series models for 53 countries. Each of these ten models is augmented with an exogenous variable—either and ad-hoc global inflation factor or Brent oil price. Overall, in almost 90% of the countries the use of any of these two variables improves the forecasting accuracy compared to the case without any exogenous factor. In 74 and 60% of the countries the global-inflation-based forecast outperforms oil-based forecast at 1- and 12-months-ahead. Twenty-four-months ahead the oil-based-forecast outperforms in 62% of the countries. Major predictive gains are observed for European OECD and Caribbean countries. |
| Date: | 2015–09 |
| URL: | http://d.repec.org/n?u=RePEc:chb:bcchwp:770&r=for |
| By: | Gospodinov, Nikolay (Federal Reserve Bank of Atlanta) |
| Abstract: | This note documents a curious finding about the substantial forecast ability of a simple aggregator of three commodity futures prices for U.S. core inflation. The proposed aggregator reduces the out-of-sample root mean squared error for 12-month-ahead inflation forecasts of the benchmark AR(1) model by 28 percent (20 percent) for the PCE (CPI) measure of core inflation. To avoid obfuscation of the sources of forecast ability, the model is intentionally kept simple, although extensions for improving and increasing the robustness of the forecast procedure are also discussed. |
| Keywords: | core inflation; commodity futures; convenience yields; forecasting |
| JEL: | C53 E37 G12 |
| Date: | 2016–02–01 |
| URL: | http://d.repec.org/n?u=RePEc:fip:fedawp:2016-05&r=for |
| By: | Barbarino, Alessandro (Board of Governors of the Federal Reserve System (U.S.)); Bura, Efstathia (George Washington University) |
| Abstract: | Factor models have been successfully employed in summarizing large datasets with few underlying latent factors and in building time series forecasting models for economic variables. When the objective is to forecast a target variable y with a large set of predictors x, the construction of the summary of the xs should be driven by how informative on y it is. Most existing methods first reduce the predictors and then forecast y in independent phases of the modeling process. In this paper we present an alternative and potentially more attractive alternative: summarizing x as it relates to y, so that all the information in the conditional distribution of y|x is preserved. These y-targeted reductions of the predictors are obtained using Sufficient Dimension Reduction techniques. We show in simulations and real data analysis that forecasting models based on sufficient reductions have the potential of significantly improved performance. |
| Keywords: | Diffusion Index; Dimension Reduction; Factor Models; Forecasting; Partial Least Squares; Principal Components |
| JEL: | C32 C53 E17 |
| Date: | 2015–09–14 |
| URL: | http://d.repec.org/n?u=RePEc:fip:fedgfe:2015-74&r=for |
| By: | Barunik, Jozef; Krehlik, Tomas; Vacha, Lukas |
| Abstract: | This paper proposes an enhanced approach to modeling and forecasting volatility using high frequency data. Using a forecasting model based on Realized GARCH with multiple time-frequency decomposed realized volatility measures, we study the influence of different timescales on volatility forecasts. The decomposition of volatility into several timescales approximates the behaviour of traders at corresponding investment horizons. The proposed methodology is moreover able to account for impact of jumps due to a recently proposed jump wavelet two scale realized volatility estimator. We propose a realized Jump-GARCH models estimated in two versions using maximum likelihood as well as observation-driven estimation framework of generalized autoregressive score. We compare forecasts using several popular realized volatility measures on foreign exchange rate futures data covering the recent financial crisis. Our results indicate that disentangling jump variation from the integrated variation is important for forecasting performance. An interesting insight into the volatility process is also provided by its multiscale decomposition. We find that most of the information for future volatility comes from high frequency part of the spectra representing very short investment horizons. Our newly proposed models outperform statistically the popular as well conventional models in both one-day and multi-period-ahead forecasting. |
| Keywords: | realized GARCH,wavelet decomposition,jumps,multi-period-ahead volatility forecasting |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:fmpwps:55&r=for |
| By: | Cláudia Duarte; Paulo M.M. Rodrigues; António Rua |
| Abstract: | The recent worldwide development and widespread use of electronic payment systems opened the opportunity to explore new data sources for monitoring macroeconomic activity. In this paper, we analyse the usefulness of data collected from Automated Teller Machines (ATM) and Points-Of-Sale (POS) for nowcasting and forecasting quarterly private consumption. To take advantage of the high frequency availability of such data, we use Mixed Data Sampling (MIDAS) regressions. A comparison of several MIDAS variants proposed in the literature is conducted, both single- and multiple variable models are considered, as well as different information sets within the quarter. Given the high penetration of ATM/POS technology in Portugal, it becomes a natural case study to assess its information content for tracking private consumption behaviour. We find that ATM/POS data displays better forecast performance than typical indicators, reinforcing the potential usefulness of this novel type of data among policymakers and practitioner. |
| JEL: | C53 E27 |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:ptu:wpaper:w201601&r=for |
| By: | Kuhns, Annemarie; Leibtag, Ephraim; Volpe, Richard; Roeger, Ed |
| Abstract: | Wholesale and retail food price forecasts are useful to farmers, processors, wholesalers, consumers, and policymakers alike, as the structure and environment of food and agricultural economies are continually evolving. USDA's Economic Research Service analyzes food prices and provides 12- to 18-month food price forecasts for 7 farm, 6 wholesale, and 19 retail food categories. In 2011, ERS’s forecasting procedure was updated to employ a vertical price transmission method that incorporates input prices at each stage of production. Where this is not possible, an autoregressive moving average approach is used. This report provides a detailed description of the revised methodology as well as an analysis of the overall accuracy and performance of individual forecasts. The revised forecasting methods show modest increases in forecast accuracy compared with simple univariate approaches previously used by ERS. |
| Keywords: | Food Price Outlook, food prices, Consumer Price Index (CPI), Producer Price Index (PPI), forecasts, vertical price transmission model, autoregressive moving average approach, error correction model, autoregressive distributed lag, univariate moving average approach, Demand and Price Analysis, |
| Date: | 2015–05 |
| URL: | http://d.repec.org/n?u=RePEc:ags:uerstb:206500&r=for |
| By: | Gospodinov, Nikolay (Federal Reserve Bank of Atlanta); Wei, Bin (Federal Reserve Bank of Atlanta) |
| Abstract: | In this paper, we examine the forecasting ability of an affine term structure framework that jointly models the markets for Treasuries, inflation-protected securities, inflation derivatives, and oil future prices based on no-arbitrage restrictions across these markets. On the methodological side, we propose a novel way of incorporating information from these markets into an affine model. On the empirical side, two main findings emerge from our analysis. First, incorporating information from inflation options can often produce more accurate inflation forecasts than those based on the Survey of Professional Forecasters. Second, incorporating oil futures tends to improve short-term inflation and longer-term nominal yield forecasts. |
| Keywords: | bond prices; TIPS; inflation derivatives; oil prices; no-arbitrage; affine models; out-of-sample forecasting |
| JEL: | C32 E43 E44 G12 |
| Date: | 2016–02–01 |
| URL: | http://d.repec.org/n?u=RePEc:fip:fedawp:2016-03&r=for |
| By: | Petre, Ionut Laurentiu |
| Abstract: | In this study I discussed about the results of the FAO studies, I used certain methods of forecasting supply and demand and Ia presented some solutions. Discussions on FAO results are based on population trends, both general and structured areas, urban and rural. Next I have predicted supply and demand for major food products in different geographic areas. These forecasts were established using economic-mathematical calculation methods. Thus, in terms of demand linear regression model was used predict it simple and to offer trend extrapolation method was used to forecast production and to predict import simple regression linear model. Thus the two main components of the market, supply and demand in each of the areas examined were put in antithesis and each represented by a graph for each main agro-food product. The study concludes with several recommendations with which it can establish a balance between supply and demand; recommendations such as reducing yield gaps, boosting the production and reducing losses. |
| Keywords: | Food security, supply and demand for agricultural products, world population, consumption and production |
| JEL: | Q1 Q11 Q18 |
| Date: | 2015–11–20 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:69269&r=for |
| By: | Carl Bonham (UH-Manoa Department of Economics, University of Hawaii Economic Research Organization); Peter Fuleky (UH-Manoa Department of Economics and University of Hawaii Economic Research Organization); James Jones (UH-Manoa Department of Economics and University of Hawaii Economic Research Organization); Ashley Hirashima (UH-Manoa Department of Economics and University of Hawaii Economic Research Organization) |
| Abstract: | We evaluate the short term forecasting performance of methods that systematically incorporate high frequency information via covariates. Our results indicate that including timely intra-period data into the forecasting process results in significant gains in predictive accuracy compared to relying exclusively on low frequency aggregates. Anticipating growing popularity of these tools among empirical analysts, we o↵er practical implementation guidelines to facilitate their adoption. |
| Keywords: | Nowcast, Ragged edge, Mixed frequency models |
| JEL: | H51 I12 Q51 Q53 |
| Date: | 2015–09 |
| URL: | http://d.repec.org/n?u=RePEc:hae:wpaper:2015-3&r=for |
| By: | Dean P. Foster; Sergiu Hart |
| Abstract: | We propose to smooth out the calibration score, which measures how good a forecaster is, by combining nearby forecasts. While regular calibration can be guaranteed only by randomized forecasting procedures, we show that smooth calibration can be guaranteed by deterministic procedures. As a consequence, it does not matter if the forecasts are leaked, i.e., made known in advance: smooth calibration can nevertheless be guaranteed (while regular calibration cannot). Moreover, our procedure has finite recall, is stationary, and all forecasts lie on a finite grid. We also consider related problems: online linear regression, weak calibration, and uncoupled Nash dynamics in n-person games. |
| Date: | 2015–09 |
| URL: | http://d.repec.org/n?u=RePEc:huj:dispap:dp692&r=for |
| By: | Stacey, Brian |
| Abstract: | Efforts to accurately predict health outcomes with a focus on informing policy makers of where to best spend limited resources have been made in the past. This paper builds on the efforts of those studies in an attempt to build an accurate predictor of health from readily available data. The American Time Use Survey (2010, 2012, and 2013) provides the majority of the data from which this model is built, and it is then tested via several methods. The analysis finds that the existing freely available data is significant in its predictive power, however is missing too many predictors to reduce the confidence interval about each individual prediction to a point of bearing meaningful fruit. That does not eliminate the usefulness of the study however, as by reducing the confidence required and accepting that the data is used for predicting societal means, the model is able to accurately predict average outcomes. This paper further attempts to analyze state level date to provide a geographic target for public funds expenditures, and accomplishes this through the analysis of various risk factors by region. Notable in this analysis is an attempt to correct for self-reporting errors. The literature review did not reveal any previous attempts to do so using a similar methodology (beyond recognizing that such errors exist and using robust methods to account for them), making this attempt possibly unique. The correction did not result in significantly different estimates, however that may be a result of the minimal resources applied to this small aspect of the analysis. |
| Keywords: | Health Outcomes, Health Econometrics, Health Prediction, Self-reporting Bias Correction |
| JEL: | C89 I14 I18 |
| Date: | 2015–11–25 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:68915&r=for |