|
on Forecasting |
| By: | Kurmas Akdogan; Selen Baser; Meltem Gulenay Chadwick; Dilara Ertug; Timur Hulagu; Sevim Kosem; Fethi Ogunc; M. Utku Ozmen; Necati Tekatli |
| Abstract: | In this paper, we produce short term forecasts for the inflation in Turkey, using a large number of econometric models. In particular, we employ univariate models, decomposition based approaches (both in frequency and time domain), a Phillips curve motivated time varying parameter model, a suite of VAR and Bayesian VAR models and dynamic factor models. Our findings suggest that the models which incorporate more economic information outperform the benchmark random walk, and the relative performance of forecasts are on average 30 percent better for the first two quarters ahead. We further combine our forecasts by means of several weighting schemes. Results reveal that, the forecast combination leads to a reduction in forecast error compared to most of the models, although some of the individual models perform alike in certain horizons. |
| Keywords: | Short-term Forecasting, Forecast Combination |
| JEL: | C52 C53 E37 |
| Date: | 2012 |
| URL: | http://d.repec.org/n?u=RePEc:tcb:wpaper:1209&r=for |
| By: | Liudas Giraitis (Queen Mary, University of London); George Kapetanios (Queen Mary, University of London); Simon Price (Bank of England and City University) |
| Abstract: | We consider time series forecasting in the presence of ongoing structural change where both the time series dependence and the nature of the structural change are unknown. Methods that downweight older data, such as rolling regressions, forecast averaging over different windows and exponentially weighted moving averages, known to be robust to historical structural change, are found to be also useful in the presence of ongoing structural change in the forecast period. A crucial issue is how to select the degree of downweighting, usually defined by an arbitrary tuning parameter. We make this choice data dependent by minimizing forecast mean square error, and provide a detailed theoretical analysis of our proposal. Monte Carlo results illustrate the methods. We examine their performance on 191 UK and US macro series. Forecasts using data-based tuning of the data discount rate are shown to perform well. |
| Keywords: | Recent and ongoing structural change, Forecast combination, Robust forecasts |
| JEL: | C10 C59 |
| Date: | 2012–03 |
| URL: | http://d.repec.org/n?u=RePEc:qmw:qmwecw:wp691&r=for |
| By: | Marcos dal Bianco (BBVA Research); Maximo Camacho (Universidad de Murcia); Gabriel Perez-Quiros (Banco de España) |
| Abstract: | We propose a fundamentals-based econometric model for the weekly changes in the euro-dollar rate with the distinctive feature of mixing economic variables quoted at different frequencies. The model obtains good in-sample fi t and, more importantly, encouraging outof-sample forecasting results at horizons ranging from one week to one month. Specifi cally, we obtain statistically signifi cant improvements upon the hard-to-beat random walk model using traditional statistical measures of forecasting error at all horizons. Moreover, our model improves greatly when we use the direction-of-change metric, which has more economic relevance than other loss measures. With this measure, our model performs much better at all forecasting horizons than a naive model that predicts the exchange rate has an equal chance to go up or down, with statistically signifi cant improvements. |
| Keywords: | euro-dollar rate, exchange rate forecasting, State-space model, mixed frequencies |
| JEL: | F31 F37 C01 C22 |
| Date: | 2012–02 |
| URL: | http://d.repec.org/n?u=RePEc:bde:wpaper:1203&r=for |
| By: | Tara M. Sinclair (George Washington University); H.O. Stekler (George Washington University); Warren Carnow (George Washington University) |
| Abstract: | This paper presents a new approach to evaluating multiple economic forecasts. In the past, evaluations have focused on the forecasts of individual variables. However, many macroeconomic variables are forecast at the same time and are used together to describe the state of the economy. It is, therefore, appropriate to examine those forecasts jointly. This specific approach is based on the Sinclair and Stekler (forthcoming) analysis of data revisions. The main contributions of this paper are (1) the application of this technique to the Survey of Professional Forecasters (SPF) and (2) showing that there is a bias that is associated with the stages of the business cycle. |
| Keywords: | Federal Reserve, Forecast Evaluation, Survey of Professional Forecasts, Business Cycle, Mahalanobis Distance |
| JEL: | C5 E2 E3 |
| Date: | 2012–03 |
| URL: | http://d.repec.org/n?u=RePEc:gwc:wpaper:2012-004&r=for |
| By: | Liudas Giraitis; George Kapetanios; Simon Price |
| Abstract: | We consider time series forecasting in the presence of ongoing structural change where both the time series dependence and the nature of the structural change are unknown. Methods that downweight older data, such as rolling regressions, forecast averaging over different windows and exponentially weighted moving averages, known to be robust to historical structural change, are found to be also useful in the presence of ongoing structural change in the forecast period. A crucial issue is how to select the degree of downweighting, usually defined by an arbitrary tuning parameter. We make this choice data dependent by minimizing forecast mean square error, and provide a detailed theoretical analysis of our proposal. Monte Carlo results illustrate the methods. We examine their performance on 191 UK and US macro series. Forecasts using data-based tuning of the data discount rate are shown to perform well. |
| Date: | 2012–03 |
| URL: | http://d.repec.org/n?u=RePEc:acb:camaaa:2012-14&r=for |
| By: | Camacho, Maximo; Pérez-Quirós, Gabriel; Poncela, Pilar |
| Abstract: | To perform real-time business cycle inferences and forecasts of GDP growth rates in the Euro area, we use an extension of the Markov-switching dynamic factor models that accounts for the specificities of the day to day monitoring of economic developments such as ragged edges, mixed frequencies and data revisions. We provide examples that show the nonlinear nature of the relations between data revisions, point forecasts and forecast uncertainty. According to our empirical results, we think that the real-time probabilities of recession inferred from the model are an appropriate statistic to capture what the press call green shoots or to monitor the double-dip recession |
| Keywords: | Business Cycles; Time Series; Turning Points |
| JEL: | C22 E27 E32 |
| Date: | 2012–03 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:8896&r=for |
| By: | Henning Fischer (University of Giessen); Marta García-Bárzana (University of Oviedo); Peter Tillmann (University of Giessen); Peter Winker (University of Giessen) |
| Abstract: | The Federal Open Market Committee (FOMC) of the U.S. Federal Reserve publishes the range of members’ forecasts for key macroeconomic variables, but not the distribution of forecasts within this range. To evaluate these projections, previous papers compare the midpoint of the ranges with the realized outcome. This paper proposes a new approach to forecast evaluation that takes account of the interval nature of projections. It is shown that using the conventional Mincer-Zarnowitz approach to evaluate FOMC forecasts misses important information contained in the width of the forecast interval. This additional information plays a minor role at short forecast horizons but turns out to be of crucial importance for inflation and unemployment forecasts 18 months into the future. At long horizons the variation of members’ projections contains information which is more relevant for explaining future inflation than information embodied in the midpoint. |
| Keywords: | Forecast evaluation, interval data, Federal Reserve, monetary policy |
| JEL: | C53 E37 E58 |
| Date: | 2012 |
| URL: | http://d.repec.org/n?u=RePEc:mar:magkse:201213&r=for |
| By: | Alvarez, Rocio; Camacho, Maximo; Pérez-Quirós, Gabriel |
| Abstract: | We examine the finite-sample performance of small versus large scale dynamic factor models. Our Monte Carlo analysis reveals that small scale factor models out-perform large scale models in factor estimation and forecasting for high levels of cross-correlation across the idiosyncratic errors of series belonging to the same category, for oversampled categories and, especially, for high persistence in either the common factor series or the idiosyncratic errors. Using a panel of 147 US economic indicators, which are classified into 13 economic categories, we show that a small scale dynamic factor model that uses one representative indicator of each category yields satisfactory or even better forecasting results than a large scale dynamic factor model that uses all the economic indicator |
| Keywords: | business cycles; output growth; time series |
| JEL: | C22 E27 E32 |
| Date: | 2012–03 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:8867&r=for |
| By: | Juan Carlos Escanciano (Indiana University); Pei Pei (Indiana University and Chinese Academy of Finance and Development, Central University of Finance and Economics) |
| Abstract: | Historical Simulation (HS) and its variant, the Filtered Historical Simulation (FHS), are the most widely used Value-at-Risk forecast methods at commercial banks. These forecast methods are traditionally evaluated by means of the unconditional backtest. This paper formally shows that the unconditional backtest is always inconsistent for backtesting HS and FHS models, with a power function that can be even smaller than the nominal level in large samples. Our ndings have fundamental implications in the determination of market risk capital requirements, and also explain Monte Carlo and empirical ndings in previous studies. We also propose a data-driven weighted backtest with good power properties to evaluate HS and FHS forecasts. Finally, our theoretical ndings are conrmed in a Monte Carlo simulation study and an empirical application with three U.S. stocks. The empirical application shows that multiplication factors computed under the current regulatory framework are downward biased, as they inherit the inconsistency of the unconditional backtest. |
| Date: | 2012–02 |
| URL: | http://d.repec.org/n?u=RePEc:inu:caeprp:2012-003&r=for |
| By: | Giannone, Domenico; Lenza, Michele; Primiceri, Giorgio E |
| Abstract: | Vector autoregressions (VARs) are flexible time series models that can capture complex dynamic interrelationships among macroeconomic variables. However, their dense parameterization leads to unstable inference and inaccurate out-of-sample forecasts, particularly for models with many variables. A potential solution to this problem is to use informative priors, in order to shrink the richly parameterized unrestricted model towards a parsimonious naïve benchmark, and thus reduce estimation uncertainty. This paper studies the optimal choice of the informativeness of these priors, which we treat as additional parameters, in the spirit of hierarchical modeling. This approach is theoretically grounded, easy to implement, and greatly reduces the number and importance of subjective choices in the setting of the prior. Moreover, it performs very well both in terms of out-of-sample forecasting, and accuracy in the estimation of impulse response functions. |
| Keywords: | Bayesian Methods; Forecasting; Hierarchical Modeling; Impulse Responses; Marginal Likelihood |
| JEL: | C11 C32 C52 E37 |
| Date: | 2012–01 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:8755&r=for |
| By: | Carriero, Andrea; Clark, Todd; Marcellino, Massimiliano |
| Abstract: | The estimation of large Vector Autoregressions with stochastic volatility using standard methods is computationally very demanding. In this paper we propose to model conditional volatilities as driven by a single common unobserved factor. This is justified by the observation that the pattern of estimated volatilities in empirical analyses is often very similar across variables. Using a combination of a standard natural conjugate prior for the VAR coefficients, and an independent prior on a common stochastic volatility factor, we derive the posterior densities for the parameters of the resulting BVAR with common stochastic volatility (BVAR-CSV). Under the chosen prior the conditional posterior of the VAR coefficients features a Kroneker structure that allows for fast estimation, even in a large system. Using US and UK data, we show that, compared to a model with constant volatilities, our proposed common volatility model significantly improves model fit and forecast accuracy. The gains are comparable to or as great as the gains achieved with a conventional stochastic volatility specification that allows independent volatility processes for each variable. But our common volatility specification greatly speeds computations. |
| Keywords: | Bayesian VARs; forecasting; prior specification; stochastic volatility |
| JEL: | C11 C13 C33 C53 |
| Date: | 2012–03 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:8894&r=for |
| By: | Chang-Jin Kim (University of Washington and Korea University); Cheolbeom Park (Department of Economics, Korea University, Seoul, Republic of Korea) |
| Abstract: | The conventional dividend-price ratio is highly persistent, and the literature reports mixed evidence on its role in predicting stock returns. In particular, its predictive power seems to be sensitive to the choice of the sample period. We argue that the decreasing number of firms with traditional dividend-payout policy is responsible for these results, and develop a model in which the long-run relationship between the dividends and stock price is time-varying. An adjusted dividend-price ratio that accounts for the time-varying long-run relationship is stationary with considerably less persistence than the conventional dividend-price ratio. Furthermore, the predictive regression model that employs the adjusted dividend-price ratio as a regressor outperforms the random-walk model in terms of long-horizon out-of-sample predictability. These results are robust with respect to the firm size. |
| Keywords: | Stock Return Predictability, Adjusted Dividend-price ratio, Disappearing,Dividends, Time-Varying Cointegration Vector, |
| JEL: | G12 C12 C22 |
| Date: | 2012 |
| URL: | http://d.repec.org/n?u=RePEc:iek:wpaper:1205&r=for |
| By: | Daziano, Ricardo A.; Achtnicht, Martin |
| Abstract: | In this paper we use Bayes estimates of a multinomial probit model with fully flexible substitution patterns to forecast consumer response to ultra-low-emission vehicles. In this empirical application of the probit Gibbs sampler, we use stated-preference data on vehicle choice from a Germany-wide survey of potential light-duty-vehicle buyers using computer-assisted personal interviewing. We show that Bayesian estimation of a multinomial probit model with a full covariance matrix is feasible for this medium-scale problem. Using the posterior distribution of the parameters of the vehicle choice model as well as the GHK simulator we derive the choice probabilities of the different alternatives. We first show that the Bayes point estimates of the market shares reproduce the observed values. Then, we define a base scenario of vehicle attributes that aims at representing an average of the current vehicle choice situation in Germany. Consumer response to qualitative changes in the base scenario is subsequently studied. In particular, we analyze the effect of increasing the network of service stations for charging electric vehicles as well as for refueling hydrogen. The result is the posterior distribution of the choice probabilities that represent adoption of the energy-effcient technologies. -- |
| Keywords: | Discrete choice models,Bayesian econometrics,Low emission vehicles,Charging infrastructure |
| JEL: | C25 D12 Q42 |
| Date: | 2012 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:zewdip:12017&r=for |
| By: | Muhammad, Shahbaz |
| Abstract: | The study investigates effect of trade openness on economic growth in the long run. We apply the ARDL bounds testing approach to test for a long run relationship and the augmented production function by incorporating financial development as an additional determinant of economic growth using the framework of Mankiw (1992). The results confirm cointegration among the series. In long run, trade openness promotes economic growth. The growth-led-trade hypothesis is vindicated by VECM Granger causality test. The causality is also checked by using the innovative accounting approach. |
| Keywords: | Trade; Growth; Cointegration; Causality; Pakistan |
| JEL: | F1 |
| Date: | 2012–01–01 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:37391&r=for |
| By: | Bhirombhakdi, Kornpob |
| Abstract: | This economic experiment initiates in evaluating a model's performance in predicting a decision. The reciprocity model is measured its accuracy rate in prediction and informativeness as aspects of the model's performance. Seventy-nine undergraduate students voluntarily joined the experiment. They made decisions contingently in designed situations as the first player in a dictator game and all roles in trust-share games. The study controls effects of choice set (equal split, competitive, and different social welfare choices) and framing effect. The result shows that the model has high performance in both prediction and informative. Furthermore, it shows an existence of the loss aversion behavior, and a significant relationship between decisions in the dictator game and the trustshare games. The study suggests that the more complicated model may not be marginally useful in predicting decision in the positive reciprocity situations. |
| Keywords: | economic experiment; performance; reciprocity; trust-share game |
| JEL: | B40 C71 C91 |
| Date: | 2011–12–23 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:37468&r=for |