Why do regime‐switching models forecast so badly?

Most non‐linear techniques give good in‐sample fits to exchange rate data but are usually outperformed by random walks or random walks with drift when used for out‐of‐sample forecasting. In the case of regime‐switching models it is possible to understand why forecasts based on the true model can have higher mean squared error than those of a random walk or random walk with drift. In this paper we provide some analytical results for the case of a simple switching model, the segmented trend model. It requires only a small misclassification, when forecasting which regime the world will be in, to lose any advantage from knowing the correct model specification. To illustrate this we discuss some results for the DM/dollar exchange rate. We conjecture that the forecasting result is more general and describes limitations to the use of switching models for forecasting. This result has two implications. First, it questions the leading role of the random walk hypothesis for the spot exchange rate. Second, it suggests that the mean square error is not an appropriate way to evaluate forecast performance for non‐linear models. Copyright © 1999 John Wiley & Sons, Ltd.     

Preliminary data and econometric forecasting: an application with the Bank of Italy Quarterly Model

This paper discusses the use of preliminary data in econometric forecasting. The standard practice is to ignore the distinction between preliminary and final data, the forecasts that do so here being termed naïve forecasts. It is shown that in dynamic models a multistep‐ahead naïve forecast can achieve a lower mean square error than a single‐step‐ahead one, as it is less affected by the measurement noise embedded in the preliminary observations. The minimum mean square error forecasts are obtained by optimally combining the information provided by the model and the new information contained in the preliminary data, which can be done within the state space framework as suggested in numerous papers. Here two simple, in general suboptimal, methods of combining the two sources of information are considered: modifying the forecast initial conditions by means of standard regressions and using intercept corrections. The issues are explored using Italian national accounts data and the Bank of Italy Quarterly Econometric Model. Copyright © 2006 John Wiley & Sons, Ltd.     

Long‐run forecasting in multicointegrated systems

We extend the analysis of Christoffersen and Diebold (1998) on long‐run forecasting in cointegrated systems to multicointegrated systems. For the forecast evaluation we consider several loss functions, each of which has a particular interpretation in the context of stock‐flow models where multicointegration typically occurs. A loss function based on a standard mean square forecast error (MSFE) criterion focuses on the forecast errors of the flow variables alone. Likewise, a loss function based on the triangular representation of cointegrated systems (suggested by Christoffersen and Diebold) considers forecast errors associated with changes in both stock (modelled through the cointegrating restrictions) and flow variables. We suggest a new loss function based on the triangular representation of multicointegrated systems which further penalizes deviations from the long‐run relationship between the levels of stock and flow variables as well as changes in the flow variables. Among other things, we show that if one is concerned with all possible long‐run relations between stock and flow variables, this new loss function entails high and increasing forecasting gains compared to both the standard MSFE criterion and Christoffersen and Diebold's criterion. This paper demonstrates the importance of carefully selecting loss functions in forecast evaluation of models involving stock and flow variables. Copyright © 2004 John Wiley & Sons, Ltd.     

Forecasting with missing data: application to coastal wave heights

This paper presents a comparative analysis of linear and mixed models for short‐term forecasting of a real data series with a high percentage of missing data. Data are the series of significant wave heights registered at regular periods of three hours by a buoy placed in the Bay of Biscay. The series is interpolated with a linear predictor which minimizes the forecast mean square error. The linear models are seasonal ARIMA models and the mixed models have a linear component and a non‐linear seasonal component. The non‐linear component is estimated by a non‐parametric regression of data versus time. Short‐term forecasts, no more than two days ahead, are of interest because they can be used by the port authorities to notify the fleet. Several models are fitted and compared by their forecasting behaviour. Copyright © 1999 John Wiley & Sons, Ltd.     

Optimal sampling frequency for volatility forecast models for the Indian stock markets

This paper evaluates the performance of conditional variance models using high‐frequency data of the National Stock Index (S&P CNX NIFTY) and attempts to determine the optimal sampling frequency for the best daily volatility forecast. A linear combination of the realized volatilities calculated at two different frequencies is used as benchmark to evaluate the volatility forecasting ability of the conditional variance models (GARCH (1, 1)) at different sampling frequencies. From the analysis, it is found that sampling at 30 minutes gives the best forecast for daily volatility. The forecasting ability of these models is deteriorated, however, by the non‐normal property of mean adjusted returns, which is an assumption in conditional variance models. Nevertheless, the optimum frequency remained the same even in the case of different models (EGARCH and PARCH) and different error distribution (generalized error distribution, GED) where the error is reduced to a certain extent by incorporating the asymmetric effect on volatility. Our analysis also suggests that GARCH models with GED innovations or EGRACH and PARCH models would give better estimates of volatility with lower forecast error estimates. Copyright © 2008 John Wiley & Sons, Ltd.     

The performance of non‐linear exchange rate models: a forecasting comparison

In recent years there has been a considerable development in modelling non‐linearities and asymmetries in economic and financial variables. The aim of the current paper is to compare the forecasting performance of different models for the returns of three of the most traded exchange rates in terms of the US dollar, namely the French franc (FF/$), the German mark (DM/$) and the Japanese yen (Y/$). The relative performance of non‐linear models of the SETAR, STAR and GARCH types is contrasted with their linear counterparts. The results show that if attention is restricted to mean square forecast errors, the performance of the models, when distinguishable, tends to favour the linear models. The forecast performance of the models is evaluated also conditional on the regime at the forecast origin and on density forecasts. This analysis produces more evidence of forecasting gains from non‐linear models. Copyright © 2002 John Wiley & Sons, Ltd.     

Signal Extraction and Forecasting of the UK Tourism Income Time Series: A Singular Spectrum Analysis Approach

We present and apply singular spectrum analysis (SSA), a relatively new, non‐parametric and data‐driven method for signal extraction (trends, seasonal and business cycle components) and forecasting of UK tourism income. Our results show that SSA slightly outperforms SARIMA and time‐varying‐parameter state space models in terms of root mean square error, mean absolute error and mean absolute percentage error forecasting criteria. Copyright © 2011 John Wiley & Sons, Ltd.     

Forecasting Volatility with Many Predictors

This study investigates the forecasting performance of the GARCH(1,1) model by adding an effective covariate. Based on the assumption that many volatility predictors are available to help forecast the volatility of a target variable, this study shows how to construct a covariate from these predictors and plug it into the GARCH(1,1) model. This study presents a method of building a covariate such that the covariate contains the maximum possible amount of predictor information of the predictors for forecasting volatility. The loading of the covariate constructed by the proposed method is simply the eigenvector of a matrix. The proposed method enjoys the advantages of easy implementation and interpretation. Simulations and empirical analysis verify that the proposed method performs better than other methods for forecasting the volatility, and the results are quite robust to model misspecification. Specifically, the proposed method reduces the mean square error of the GARCH(1,1) model by 30% for forecasting the volatility of S&P 500 Index. The proposed method is also useful in improving the volatility forecasting of several GARCH‐family models and for forecasting the value‐at‐risk. Copyright © 2013 John Wiley & Sons, Ltd.     

Realized Volatility Forecast of Stock Index Under Structural Breaks

We investigate the realized volatility forecast of stock indices under the structural breaks. We utilize a pure multiple mean break model to identify the possibility of structural breaks in the daily realized volatility series by employing the intraday high‐frequency data of the Shanghai Stock Exchange Composite Index and the five sectoral stock indices in Chinese stock markets for the period 4 January 2000 to 30 December 2011. We then conduct both in‐sample tests and out‐of‐sample forecasts to examine the effects of structural breaks on the performance of ARFIMAX‐FIGARCH models for the realized volatility forecast by utilizing a variety of estimation window sizes designed to accommodate potential structural breaks. The results of the in‐sample tests show that there are multiple breaks in all realized volatility series. The results of the out‐of‐sample point forecasts indicate that the combination forecasts with time‐varying weights across individual forecast models estimated with different estimation windows perform well. In particular, nonlinear combination forecasts with the weights chosen based on a non‐parametric kernel regression and linear combination forecasts with the weights chosen based on the non‐negative restricted least squares and Schwarz information criterion appear to be the most accurate methods in point forecasting for realized volatility under structural breaks. We also conduct an interval forecast of the realized volatility for the combination approaches, and find that the interval forecast for nonlinear combination approaches with the weights chosen according to a non‐parametric kernel regression performs best among the competing models. Copyright © 2014 John Wiley & Sons, Ltd.     

A markup model for forecasting inflation for the euro area

We develop a small model for forecasting inflation for the euro area using quarterly data over the period June 1973 to March 1999. The model is used to provide inflation forecasts from June 1999 to March 2002. We compare the forecasts from our model with those derived from six competing forecasting models, including autoregressions, vector autoregressions and Phillips‐curve based models. A considerable gain in forecasting performance is demonstrated using a relative root mean squared error criterion and the Diebold–Mariano test to make forecast comparisons. Copyright © 2006 John Wiley & Sons, Ltd.     

Forecasting the price of crude oil via convenience yield predictions

The paper develops an oil price forecasting technique which is based on the present value model of rational commodity pricing. The approach suggests shifting the forecasting problem to the marginal convenience yield, which can be derived from the cost‐of‐carry relationship. In a recursive out‐of‐sample analysis, forecast accuracy at horizons within one year is checked by the root mean squared error as well as the mean error and the frequency of a correct direction‐of‐change prediction. For all criteria employed, the proposed forecasting tool outperforms the approach of using futures prices as direct predictors of future spot prices. Vis‐à‐vis the random‐walk model, it does not significantly improve forecast accuracy but provides valuable statements on the direction of change. Copyright © 2007 John Wiley & Sons, Ltd.     

Local to unity,long‐horizon forecasting thresholds for model selection in the AR(1)

This article introduces a novel framework for analysing long‐horizon forecasting of the near non‐stationary AR(1) model. Using the local to unity specification of the autoregressive parameter, I derive the asymptotic distributions of long‐horizon forecast errors both for the unrestricted AR(1), estimated using an ordinary least squares (OLS) regression, and for the random walk (RW). I then identify functions, relating local to unity ‘drift’ to forecast horizon, such that OLS and RW forecasts share the same expected square error. OLS forecasts are preferred on one side of these ‘forecasting thresholds’, while RW forecasts are preferred on the other. In addition to explaining the relative performance of forecasts from these two models, these thresholds prove useful in developing model selection criteria that help a forecaster reduce error. Copyright © 2004 John Wiley & Sons, Ltd.     

The importance of non-linearities in large forecasting models with stochastic error processes

This paper considers the consequences of the stochastic error process in large non-linear forecasting models. As such models are non-linear, the deterministic forecast is neither the mean nor the mode of the density function of the endogenous variables. Under a specific assumption as to the class of the non-linearity it is shown that the deterministic forecast is actually the vector of marginal medians of the density function. Stochastic simulation techniques are then used to test whether one large forecasting model actually lies within this class.     

When are Direct Multi‐step and Iterative Forecasts Identical?

Although both direct multi‐step‐ahead forecasting and iterated one‐step‐ahead forecasting are two popular methods for predicting future values of a time series, it is not clear that the direct method is superior in practice, even though from a theoretical perspective it has lower mean squared error (MSE). A given model can be fitted according to either a multi‐step or a one‐step forecast error criterion, and we show here that discrepancies in performance between direct and iterative forecasting arise chiefly from the method of fitting, and is dictated by the nuances of the model's misspecification. We derive new formulas for quantifying iterative forecast MSE, and present a new approach for assessing asymptotic forecast MSE. Finally, the direct and iterative methods are compared on a retail series, which illustrates the strengths and weaknesses of each approach. Copyright © 2015 John Wiley & Sons, Ltd.     

Forecasting US employment growth using forecast combining methods

We examine different approaches to forecasting monthly US employment growth in the presence of many potentially relevant predictors. We first generate simulated out‐of‐sample forecasts of US employment growth at multiple horizons using individual autoregressive distributed lag (ARDL) models based on 30 potential predictors. We then consider different methods from the extant literature for combining the forecasts generated by the individual ARDL models. Using the mean square forecast error (MSFE) metric, we investigate the performance of the forecast combining methods over the last decade, as well as five periods centered on the last five US recessions. Overall, our results show that a number of combining methods outperform a benchmark autoregressive model. Combining methods based on principal components exhibit the best overall performance, while methods based on simple averaging, clusters, and discount MSFE also perform well. On a cautionary note, some combining methods, such as those based on ordinary least squares, often perform quite poorly. Copyright © 2008 John Wiley & Sons, Ltd.     

How helpful are spatial effects in forecasting the growth of Chinese provinces?

In this paper, we make multi‐step forecasts of the annual growth rates of the real gross regional product (GRP) for each of the 31 Chinese provinces simultaneously. Beside the usual panel data models, we use panel models that explicitly account for spatial dependence between the GRP growth rates. In addition, the possibility of spatial effects being different for different groups of provinces (Interior and Coast) is allowed for. We find that both pooling and accounting for spatial effects help substantially to improve the forecast performance compared to the benchmark models estimated for each of the provinces separately. It is also shown that the effect of accounting for spatial dependence is even more pronounced at longer forecasting horizons (the forecast accuracy gain as measured by the root mean squared forecast error is about 8% at the 1‐year horizon and exceeds 25% at the 13‐ and 14‐year horizons). Copyright © 2010 John Wiley & Sons, Ltd.     

Forecasting German GDP using alternative factor models based on large datasets

This paper discusses the forecasting performance of alternative factor models based on a large panel of quarterly time series for the German economy. One model extracts factors by static principal components analysis; the second model is based on dynamic principal components obtained using frequency domain methods; the third model is based on subspace algorithms for state‐space models. Out‐of‐sample forecasts show that the forecast errors of the factor models are on average smaller than the errors of a simple autoregressive benchmark model. Among the factor models, the dynamic principal component model and the subspace factor model outperform the static factor model in most cases in terms of mean‐squared forecast error. However, the forecast performance depends crucially on the choice of appropriate information criteria for the auxiliary parameters of the models. In the case of misspecification, rankings of forecast performance can change severely. Copyright © 2007 John Wiley & Sons, Ltd.     

Linear constraints,robust-weighting and efficient composite modeling

Recent studies have shown that composite forecasting produces superior forecasts when compared to individual forecasts. This paper extends the existing literature by employing linear constraints and robust regression techniques in composite model building. Security analysts forecasts may be improved when combined with time series forecasts for a diversified sample of 261 firms with a 1980-1982 post-sample estimation period. The mean square error of analyst forecasts may be reduced by combining analyst and univariate time series model forecasts in constrained and unconstrained ordinary least squares regression models. These reductions are very interesting when one finds that the univariate time series model forecasts do not substantially deviate from those produced by ARIMA (0,1,1) processes. Moreover, security analysts' forecast errors may be significantly reduced when constrained and unconstrained robust regression analyses are employed.     

A hybrid forecasting approach for piece‐wise stationary time series

We consider the problem of forecasting a stationary time series when there is an unknown mean break close to the forecast origin. Based on the intercept‐correction methods suggested by Clements and Hendry (1998) and Bewley (2003), a hybrid approach is introduced, where the break and break point are treated in a Bayesian fashion. The hyperparameters of the priors are determined by maximizing the marginal density of the data. The distributions of the proposed forecasts are derived. Different intercept‐correction methods are compared using simulation experiments. Our hybrid approach compares favorably with both the uncorrected and the intercept‐corrected forecasts. Copyright © 2006 John Wiley & Sons, Ltd.     

Forecasting with growth curves: the effect of error structure

"The main theme of this paper is an investigation into the importance of error structure as a determinant of the forecasting accuracy of the logistic model. The relationship between the variance of the disturbance term and forecasting accuracy is examined empirically. A general local logistic model is developed as a vehicle to be used in this investigation. Some brief comments are made on the assumptions about error structure, implicit or explicit, in the literature." The results suggest that "the variance of the disturbance term, when using the logistic to forecast human populations, is proportional to at least the square of population size."