Forecasting Competition Research Articles

New Insights on the Implied and Realized Volatility Relation

Existing studies on the informational content of at-the-money implied volatility (ATMIV) and past realized volatility (PRV) and the relation between the two have mainly focused on a single short forecast horizon and conclude that ATMIV outperforms PRV. We examine the relation between implied and realized volatility over both short and longer forecasting horizons to provide a forecasting competition. We analytically demonstrate the option maturity effect on the sensitivity of the implied volatility (IV) estimation. As time to maturity increases, vega increases but at a decreasing rate, up to [Formula: see text]. At shorter (longer) maturities, small pricing changes should have greater (smaller) corrective impact on IVs. We find that IV outperforms PRV over a one month forecast horizon. However, as the forecast horizon increases, PRV outperforms IV and subsumes the information contained in it. These mixed results may be attributed to the reduced efficiency of longer dated sections of the S&P 500 options index market we analyze.

Hybridization of the probabilistic neural networks with feed-forward neural networks for forecasting

Feed-forward neural networks (FFNNs) are among the most important neural networks that can be applied to a wide range of forecasting problems with a high degree of accuracy. Several large-scale forecasting competitions with a large number of commonly used time series forecasting models conclude that combining forecasts from more than one model often leads to improved performance, especially when the models in the ensemble are quite different. In the literature, several hybrid models have been proposed by combining different time series models together. In this paper, in contrast of the traditional hybrid models, a novel hybridization of the feed-forward neural networks (FFNNs) is proposed using the probabilistic neural networks (PNNs) in order to yield more accurate results than traditional feed-forward neural networks. In the proposed model, the estimated values of the FFNN models are modified based on the distinguished trend of their residuals and optimum step length, which are respectively yield from a probabilistic neural network and a mathematical programming model. Empirical results with three well-known real data sets indicate that the proposed model can be an effective way in order to construct a more accurate hybrid model than FFNN models. Therefore, it can be applied as an appropriate alternative model for forecasting tasks, especially when higher forecasting accuracy is needed.

Commentary on the Makridakis Time Series Competition (M-Competition)

In 1982, the Journal of Forecasting published the results of a forecasting competition organized by Spyros Makridakis (Makridakis et al., 1982). In this, the ex ante forecast errors of 21 methods were compared for forecasts of a variety of economic time series, generally using 1001 time series. Only extrapolative methods were used, as no data were available on causal variables. The accuracies of methods were compared using a variety of accuracy measures for different types of data and for varying forecast horizons.The original paper did not contain much interpretation or discussion. Partly this was by design, to be unbiased in the presentation. A more important factor, however, was the difficulty in gaining consensus on interpretation and presentation among the diverse group of authors, many of whom have a vested interest in certain methods.In the belief that this study was of major importance, we decided to obtain a more complete discussion of the results. We do not believe that “the data speak for themselves.”

The Wisdom of Competitive Crowds

When several individuals are asked to forecast an uncertain quantity, they often face implicit or explicit incentives to be the most accurate. Despite the desire to elicit honest forecasts, such competition induces forecasters to report strategically and non-truthfully. The question we address is whether the competitive crowd's forecast (the average of strategic forecasts) is more accurate than the truthful crowd's forecast (the average of truthful forecasts from the same forecasters). We analyze a forecasting competition in which a prize is awarded to the forecaster whose point forecast is closest to the actual outcome. Before reporting a forecast, we assume each forecaster receives two signals: one common and one private. These signals represent the forecasters' past shared and personal experiences relevant for forecasting the uncertain quantity of interest. In a set of equilibrium results, we characterize the nature of the strategic forecasts in this game. As the correlation among the forecasters' private signals increases, the forecasters switch from using a pure to a mixed strategy. In both cases, forecasters exaggerate their private information and thereby make the competitive crowd's forecast more accurate than the truthful crowd's forecast.

The Wisdom of Competitive Crowds

When several individuals are asked to forecast an uncertain quantity, they often face implicit or explicit incentives to be the most accurate. Despite the desire to elicit honest forecasts, such competition induces forecasters to report strategically and nontruthfully. The question we address is whether the competitive crowd's forecast (the average of strategic forecasts) is more accurate than the truthful crowd's forecast (the average of truthful forecasts from the same forecasters). We analyze a forecasting competition in which a prize is awarded to the forecaster whose point forecast is closest to the actual outcome. Before reporting a forecast, we assume each forecaster receives two signals: one common and one private. These signals represent the forecasters' past shared and personal experiences relevant for forecasting the uncertain quantity of interest. In a set of equilibrium results, we characterize the nature of the strategic forecasts in this game. As the correlation among the forecasters' private signals increases, the forecasters switch from using a pure to a mixed strategy. In both cases, forecasters exaggerate their private information and thereby make the competitive crowd's forecast more accurate than the truthful crowd's forecast.

The tourism forecasting competition

Abstract We evaluate the performances of various methods for forecasting tourism data. The data used include 366 monthly series, 427 quarterly series and 518 annual series, all supplied to us by either tourism bodies or academics who had used them in previous tourism forecasting studies. The forecasting methods implemented in the competition are univariate and multivariate time series approaches, and econometric models. This forecasting competition differs from previous competitions in several ways: (i) we concentrate on tourism data only; (ii) we include approaches with explanatory variables; (iii) we evaluate the forecast interval coverage as well as the point forecast accuracy; (iv) we observe the effect of temporal aggregation on the forecasting accuracy; and (v) we consider the mean absolute scaled error as an alternative forecasting accuracy measure. We find that pure time series approaches provide more accurate forecasts for tourism data than models with explanatory variables. For seasonal data we implement three fully automated pure time series algorithms that generate accurate point forecasts, and two of these also produce forecast coverage probabilities which are satisfactorily close to the nominal rates. For annual data we find that Naive forecasts are hard to beat.

Forecasting business and consumer surveys indicators–a time-series models competition

The objective of this article is to compare different time-series methods for the short-run forecasting of Business and Consumer Survey Indicators. We consider all available data taken from the Business and Consumer Survey Indicators for the Euro area between 1985 and 2002. The main results of the forecast competition are offered not only for raw data but we also consider the effects of seasonality and removing outliers on forecast accuracy. In most cases, the univariate autoregressions were not outperformed by the other methods. As for the effect of seasonal adjustment methods and the use of data from which outliers have been removed, we obtain that the use of raw data has little effect on forecast accuracy. The forecasting performance of qualitative indicators is important since enlarging the observed time series of these indicators with forecast intervals may help in interpreting and assessing the implications of the current situation and can be used as an input in quantitative forecast models.

Forecasting Tourism-Generated Employment: The Case of Denmark

The empirical results from a forecasting competition show that the unrestricted vector autoregressive model is likely to generate the most accurate forecasts of international tourist expenditure in Denmark. This model is therefore estimated (using data for 1969–99) and is used to generate tourism expenditure forecasts for Denmark to 2010. The employment requirements (direct, indirect and induced) associated with these expenditure forecasts are then estimated using an input–output model. The forecasts of employment demands are shown across all industrial sectors, and linked to qualifications data in respect of the labour force. The major impacts of foreign tourist expenditure on employment in Denmark occur in the retail, hotel and restaurant sectors. Foreign tourist expenditure is also significantly associated with graduate employment.

Forecasting unemployment

Results are reported of a forecasting competition between linear autoregressive, GARCH, threshold autoregressive and neural network models of the UK monthly unemployment rate series. Evidence is uncovered suggesting the existence of nonlinearities in the series.

Neural Network Models for Time Series Forecasts

Neural networks have been advocated as an alternative to traditional statistical forecasting methods. In the present experiment, time series forecasts produced by neural networks are compared with forecasts from six statistical time series methods generated in a major forecasting competition (Makridakis et al. [Makridakis, S., A. Anderson, R. Carbone, R. Fildes, M. Hibon, R. Lewandowski, J. Newton, E. Parzen, R. Winkler. 1982. The accuracy of extrapolation (time series) methods: Results of a forecasting competition. J. Forecasting 1 111–153.]); the traditional method forecasts were estimated by experts in the particular technique. The neural networks were estimated using the same ground rules as the competition. Across monthly and quarterly time series, the neural networks did significantly better than traditional methods. As suggested by theory, the neural networks were particularly effective for discontinuous time series.

Connectionist approach to time series prediction: an empirical test

Among the various potential applications of neural networks, forecasting is considered to be a major application. Several researchers have reported their experiences with the use of neural networks in forecasting, and the evidence is inconclusive. This paper presents the results of a forecasting competition between a neural network model and a Box-Jenkins automatic forecasting expert system. Seventy-five series, a subset of data series which have been used for comparison of various forecasting techniques, were analysed using the Box-Jenkins approach and a neural network implementation. The results show that the simple neural net model tested on this set of time series could forecast about as well as the Box-Jenkins forecasting system.

Results of a Price Forecasting Competition

ployed by each are described in detail in preceding papers and are only briefly described here. Berck and Chalfant employed a statistical package shown as ACE which, when given a dependent variable, y, and a set of independent variables, X, searches for the 4 = g(y) and 0 = h(X) that maximizes the correlation coefficient between 0 and 4. It then uses these transformations of the variables to forecast 4 and recover a forecast of y by applying the inverse transformation 9 = g-'(4). To forecast the price of feed, Berck and Chalfant included one-period and two-period lagged values of the price of feed deflated by the price of other goods and real per capita income lagged three times as independent variables. They used the deflated price of feed as the dependent variable, then recovered it with an AR(1) prediction of the price of other goods. For the forecasts of the price of livestock, Berck and Chalfant used seven independent variables: the deflated price of feed lagged once and four periods, per capita consumption of livestock lagged one, two, and four periods, lagged real per capita income, and the lagged price of other goods. They did not deflate the price of livestock.

A Price Forecasting Competition: Introduction

Accurate forecasts are important to professional agricultural economists. Forecasts are used not only for predicting random variables but as proxies for unobservable expectations in many analytical settings. When using forecasts, the analyst strives to obtain the most accurate forecast possible, where accuracy is determined by the specific objective of the forecasting exercise. Forecasts can be obtained by (a) purely judgmental approaches, (b) structural econometric models, (c) univariate time-series models, (d) multivariate time-series models, or (e) any combination of the above. The problem becomes one of choosing the most appropriate method (or combination of methods) for a given application. Professional agricultural economists need to understand various forecasting approaches and how they differ from one another. Such information would aid forecast users in making rational choices regarding which forecasting methods to employ for a given application. A forecasting method may not exist in all situations. Any one of the methods (or combination of the methods) mentioned above may provide the best forecasts in a given situation according to a particular loss function. Makridakas et al. summarize the conclusions from

Forecasting with balanced state space representations of multivariate distributed lag models

AbstractA procedure for estimating state space models for multivariate distributed lag processes is described. It involves singular value decomposition techniques and yields an internally balanced state space representation which has attractive properties. Following the specifications of a forecasting competition, the approach is applied to generate ex‐post forecasts for US real GNP growth rates. The forecasts of the estimated state space model are compared to those of twelve econometric models and an ARIMA model.

Forecasting and loss functions

This paper considers two problems of interpreting forecasting competition error statistics. The first problem is concerned with the importance of linking the error measure (loss function) used in evaluating a forecasting model with the loss function used in estimating the model. It is argued that because the variety of uses of any single forecast, such matching is impractical. Secondly, there is little evidence that matching would have any impact on comparative forecast performance, however measured. As a consequence the results of forecasting competitions are not affected by this problem. The second problem is concerned with the interpreting performance, when evaluated through M(ean) S(quare) E(rror). The authors show that in the Makridakis Competition, good MSE performance is solely due to performance on a small number of the 1001 series, and arises because of the effects of scale. They conclude that comparisons of forecasting accuracy based on MSE are subject to major problems of interpretation.

ARARMA vs ARIMA—A study of the benefits of a new approach to forecasting

The ARARMA methodology of time series forecasting introduced by Parzen has compared well with longer established techniques such as Box and Jenkins ARIMA models in the results of a major forecasting competition. The two main differences between these methodologies are the way data is transformed to stationarity, and the interpretation of the concept of parsimonious models. The additional benefits familiarity with the ARARMA approach can bestow on those used to the ARIMA approach are identified. Following a description and demonstration of the ARARMA methodology, a comparison of forecasting performance is made. The comparison is carried out on sets of data for which published ARIMA models are available. To ensure that the comparison is of genuine forecasting ability (rather than fitting ability), a portion of data is saved solely for forecasting performance measurement. The results indicate that there are additional benefits to be gained from the ARARMA approach. The benefits lie in the approach's tendency to avoid over differencing and in the diagnostic tools for identifying the stationary ARIMA model, rather than in the different transformation to stationarity.

The accuracy of extrapolation (time series) methods: Results of a forecasting competition

AbstractIn the last few decades many methods have become available for forecasting. As always, when alternatives exist, choices need to be made so that an appropriate forecasting method can be selected and used for the specific situation being considered. This paper reports the results of a forecasting competition that provides information to facilitate such choice. Seven experts in each of the 24 methods forecasted up to 1001 series for six up to eighteen time horizons. The results of the competition are presented in this paper whose purpose is to provide empirical evidence about differences found to exist among the various extrapolative (time series) methods used in the competition.