|
on Forecasting |
| By: | Laura Carabotta (Facultat d'Economia i Empresa; Universitat de Barcelona (UB)); Peter Claeys (Université libre de Bruxelles) |
| Abstract: | Budget forecasts have become increasingly important as a tool of fiscal management to influence expectations of bond markets and the public at large. The inherent difficulty in projecting macroeconomic variables – together with political bias – thwart the accuracy of budget forecasts. We improve accuracy by combining the forecasts of both private and public agencies for Italy over the period 1993-2012. A weighted combined forecast of the deficit/ ratio is superior to any single forecast. Deficits are hard to predict due to shifting economic conditions and political events. We test and compare predictive accuracy over time and although a weighted combined forecast is robust to breaks, there is no significant improvement over a simple RW model. |
| Keywords: | deficit, forecast accuracy, fiscal forecasting, forecast comparison,forecast combination, fluctuation test. |
| JEL: | G12 C14 E43 E62 H62 H63 |
| Date: | 2015 |
| URL: | http://d.repec.org/n?u=RePEc:ewp:wpaper:320web&r=for |
| By: | Linlin Niu; Xiu Xu; Ying Chen; |
| Abstract: | We propose the use of a local autoregressive (LAR) model for adaptive estimation and forecasting of three of China’s key macroeconomic variables: GDP growth, inflation and the 7-day interbank lending rate. The approach takes into account possible structural changes in the data-generating process to select a local homogeneous interval for model estimation, and is particularly well-suited to a transition economy experiencing ongoing shifts in policy and structural adjustment. Our results indicate that the proposed method outperforms alternative models and forecast methods, especially for forecast horizons of 3 to 12 months. Our 1-quarter ahead adaptive forecasts even match the performance of the well-known CMRC Langrun survey forecast. The selected homogeneous intervals indicate gradual changes in growth of industrial production driven by constant evolution of the real economy in China, as well as abrupt changes in interestrate and inflation dynamics that capture monetary policy shifts. |
| Keywords: | Chinese economy, local parametric models, forecasting |
| JEL: | E43 E47 |
| Date: | 2015–04 |
| URL: | http://d.repec.org/n?u=RePEc:hum:wpaper:sfb649dp2015-023&r=for |
| By: | Henzel, Steffen; Lehmann, Robert; Wohlrabe, Klaus |
| Abstract: | We tackle the nowcasting problem at the regional level using a large set of indicators (regional, national and international) for the years 1998 to 2013. We explicitly use the ragged-edge data structure and consider the different information sets faced by a regional forecaster within each quarter. It appears that regional survey results in particular improve forecasting accuracy. Among the 10% best performing models for the short forecasting horizon, one fourth contain regional indicators. Hard indicators from the German manufacturing sector and the Composite Leading Indicator for Europe also deliver useful information for the prediction of regional GDP in Saxony. Unlike national GDP forecasts, the performance of regional GDP is similar across different information sets within a quarter. |
| Keywords: | nowcasting, regional gross domestic product, bridge equations, regional economic forecasting, mixed frequency |
| JEL: | C22 C52 C53 E37 R11 |
| Date: | 2015–04–26 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:63714&r=for |
| By: | Francesco Ravazzolo; Joaquin L. Vespignani |
| Abstract: | In modelling macroeconomic time series, often a monthly indicator of global real economic activity is used. We propose a new indicator, named World steel production, and compare it to other existing indicators, precisely the Kilian’s index of global real economic activity and the index of OECD World industrial production. We develop an econometric approach based on desirable econometric properties in relation to the quarterly measure of World or global gross domestic product to evaluate and to choose across different alternatives. The method is designed to evaluate short-term, long-term and predictability properties of the indicators. World steel production is proven to be the best monthly indicator of global economic activity in terms of our econometric properties. Kilian’s index of global real economic activity also accurately predicts World GDP growth rates. When extending the analysis to an out-of-sample exercise, both Kilian’s index of global real economic activity and the World steel production produce accurate forecasts for World GDP, confirming evidence provided by the econometric properties. Specifically, a forecast combination of the three indices produces statistically significant gains up to 40% at nowcast and more than 10% at longer horizons relative to an autoregressive benchmark. |
| Keywords: | Global real economic activity, World steel production, Forecasting |
| JEL: | E1 E3 C1 C5 C8 |
| Date: | 2015–04 |
| URL: | http://d.repec.org/n?u=RePEc:een:camaaa:2015-13&r=for |
| By: | Ralph D. Snyder; J. Keith Ord; Anne B. Koehler; Keith R. McLaren; Adrian Beaumont |
| Abstract: | A method is proposed for forecasting composite time series such as the market shares for multiple brands. Its novel feature is that it relies on multi-series adaptations of exponential smoothing combined with the log-ratio transformation for the conversion of proportions onto the real line. It is designed to produce forecasts that are both non-negative and sum to one; are invariant to the choice of the base series in the log-ratio transformation; recognized and exploit features such as serial dependence and non-stationary movements in the data; allow for the possibility of non-uniform interactions between the series; and contend with series that start late, finish early, or which have values close to zero. Relying on an appropriate multivariate innovations state space mode, it can be used to generate prediction distributions in addition to point forecasts and to compute the probabilities of market share increases together with prediction intervals. A shared structure between the series in the multivariate model is used to ensure that the usual proliferation of parameter is avoided. The forecasting method is illustrated using data on the annual market shares of the major (groups of) brands in the U.S. automobile market, over the period 1961-2013. |
| Keywords: | Exponential smoothing; Proportions; Prediction intervals; Automobile sales; Market shares. |
| Date: | 2015 |
| URL: | http://d.repec.org/n?u=RePEc:msh:ebswps:2015-11&r=for |
| By: | Didenko, Alexander; Dubovikov, Michael; Poutko, Boris |
| Abstract: | The paper develops an algorithm for making long-term (up to three months ahead) predictions of volatility reversals based on long memory properties of financial time series. The approach for computing fractal dimension using sequence of the minimal covers with decreasing scale is used to decompose volatility into two dynamic components: specific and structural. We introduce two separate models for both, based on different principles and capable of catching long uptrends in volatility. To test statistical significance of its abilities we introduce several estimators of conditional and unconditional probabilities of reversals in observed and predicted dynamic components of volatility. Our results could be used for forecasting points of market transition to an unstable state. |
| Keywords: | stock market; price risk; fractal dimension; market crash; ARCH-GARCH; range-based volatility models; multi-scale volatility; volatility reversals; technical analysis. |
| JEL: | C14 C49 C5 C58 |
| Date: | 2015–03 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:63708&r=for |
| By: | Christoph Bergmeir; Rob J Hyndman; Bonsoo Koo |
| Abstract: | One of the most widely used standard procedures for model evaluation in classification and regression is K-fold cross-validation (CV). However, when it comes to time series forecasting, because of the inherent serial correlation and potential non-stationarity of the data, its application is not straightforward and often omitted by practitioners in favor of an out-of-sample (OOS) evaluation. In this paper, we show that the particular setup in which time series forecasting is usually performed using Machine Learning methods renders the use of standard K-fold CV possible. We present theoretical insights supporting our arguments. Furthermore, we present a simulation study where we show empirically that K-fold CV performs favourably compared to both OOS evaluation and other time-series-specific techniques such as non-dependent cross-validation. |
| Keywords: | cross-validation, time series, auto regression. |
| JEL: | C52 C53 C22 |
| Date: | 2015 |
| URL: | http://d.repec.org/n?u=RePEc:msh:ebswps:2015-10&r=for |
| By: | Marcelle Chauvet; Elcyon C. R. Lima; Brisne Vasquez |
| Abstract: | This paper compares the forecasting performance of linear and nonlinear models under the presence of structural breaks for the Brazilian real GDP growth. The Markov switching models proposed by Hamilton (1989) and its generalized version by Lam (1990) are applied to quarterly GDP from 1975:1 to 2000:2 allowing for breaks at the Collor Plans. The probabilities of recessions are used to analyze the Brazilian business cycle. The ability of each model in forecasting out-of-sample the growth rates of GDP is examined. The forecasting ability of the two models is also compared with linear specifications. We find that nonlinear models display the best forecasting performance and that specifications including the presence of structural breaks are important in obtaining a representation of the Brazilian business cycle. Neste artigo são comparadas as habilidades preditivas de modelos lineares e nãolineares, com quebras estruturais, para a taxa de crescimento do PIB do Brasil. São estimados os modelos com mudança de regime markoviana propostos por Hamilton (1989) e Lam (1990) - que generaliza o modelo de Hamilton - com dados trimestrais de 1975:1 a 2000:2. Na estimação dos modelos são permitidas quebras estruturais durante os Planos Collor I e II. As probabilidades de recessão dos modelos são utilizadas para se analisar o ciclo de negócios brasileiro. É examinada a capacidade de se prever a taxa de crescimento do PIB fora da amostra e a habilidade preditiva dos dois modelos é comparada com a de modelos lineares. Os nossos resultados revelam que os modelos não-lineares são os que apresentam o melhor desempenho preditivo e que a inclusão de quebras estruturais é importante para se obter a representação do ciclo de negócios no Brasil. |
| Date: | 2015–01 |
| URL: | http://d.repec.org/n?u=RePEc:ipe:ipetds:0118&r=for |
| By: | Olivier Gervais; Marc-André Gosselin |
| Abstract: | The authors describe the key features of a new large-scale Canadian macroeconomic forecasting model developed over the past two years at the Bank of Canada. The new model, called LENS for Large Empirical and Semi-structural model, uses a methodology similar to the Federal Reserve Board’s FRB/US model and the Bank of Canada’s projection model of the U.S. economy (MUSE). LENS is based on a system of estimated reduced-form equations that describe the interactions among key macroeconomic variables. The model strikes a balance between theoretical structure and empirical properties, since most behavioural equations combine forward-looking expectations with adjustment costs. Compared to ToTEM, the Bank’s main model for projection and policy analysis, LENS is more driven by the empirical properties of the data than economic theory and generally provides better out-of-sample forecast performance. In addition, LENS is more disaggregated, thereby allowing the analysis of a broader set of issues related to the economic outlook. These properties will make LENS a useful complement to ToTEM for constructing economic projections at the Bank of Canada. |
| Keywords: | Economic models; Econometric and statistical methods |
| JEL: | E37 C53 E17 E27 F17 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:bca:bocatr:102&r=for |
| By: | Goodness C. Aye (goodness.aye@gmail.com); Mehmet Balcilar (mehmet@mbalcilar.net); Rangan Gupta (Department of Economics, University of Pretoria) |
| Abstract: | This study investigates the predictability of 11 industrialized stock returns with emphasis on the role of U.S. returns. Using monthly data spanning 1980:2 to 2014:12, we show that there exist multiple structural breaks and nonlinearities in the data. Therefore, we employ methods that are capable of accounting for these and at the same time date stamping the periods of causal relationship between the U.S. returns and those of the other countries. First we implement a subsample analysis which relies on the set of models, data set and sample range as in Rapach et al. (2013). Our results show that while the U.S. returns played a strong predictive role based on the OLS pairwise Granger causality predictive regression and news-diffusion models, it played no role based on the pooled version of the OLS model and its role based on the adaptive elastic net model is weak relative to Switzerland. Second, we implement our preferred model: a bootstrap rolling window approach using our newly updated data on stock returns for each countries, and find that U.S. stock return has significant predictive ability for all the countries at certain sub-periods. Given these results, it would be misleading to rely on results based on constant-parameter linear models that assume that the relationship between the U.S. returns and those of other industrialized countries are permanent, since the relationship is, in fact, time-varying, and holds only at specific periods. |
| Keywords: | Stock returns, predictability, structural breaks, nonlinearity, time varying causality |
| JEL: | C32 G10 G15 |
| Date: | 2015–04 |
| URL: | http://d.repec.org/n?u=RePEc:pre:wpaper:201524&r=for |
| By: | Marco S. Matsumura; Ajax R. B. Moreira |
| Abstract: | The evolution of the yields of different maturities is related and can be described by a reduced number of commom latent factors. Multifactor interest rate models of the finance literature, common factor models of the time series literature and others use this property. Each model has advantages and disadvantages, and it is an empirical matter to evaluate the performance of the approaches. This exercise compares 4 alternative models for the term structure using 3 different markets: the Brazilian domestic and sovereign market and the US market. A evolução das diversas maturidades das taxas de juros está relacionada e pode ser descrita por um número reduzido de variáveis latentes comuns. Os modelos de taxas de juros multivariados da literatura de finanças utilizam esta propriedade, assim como os modelos de fator comum da literatura de séries temporais, e modelos de decomposição da curva de juros. Cada um desses modelos tem vantagens e desvantagens, sendo uma questão empírica avaliar o desempenho dessas abordagens. Esse exercício compara a resposta de quatro modelos alternativos para a curva de juros, em três mercados diferentes: juros domésticos brasileiros, juros soberanos externos brasileiros, e juros domésticos dos Estados Unidos. |
| Date: | 2015–01 |
| URL: | http://d.repec.org/n?u=RePEc:ipe:ipetds:0174&r=for |