A NEW METHOD FOR THE ESTIMATION OF CAPACITY UTILIZATION: THEORY AND EMPIRICAL EVIDENCE FROM 14 EU COUNTRIES

In this paper, we develop a new and at the same time simple method of obtaining a measure of the rate of capacity utilization (CU) which makes use of the structural vector autoregression (SVAR) system of equation estimating technique with long‐run restrictions. The measure of CU that we derive for each of 14 EU countries replicates to a great extent the European Commission's Directorate General for Economic and Financial Affairs (ECFIN) measure. On closer examination we find that the in‐sample explanatory content with respect to the inflation rate of the SVAR measure exceeds more often than not that of the ECFIN's measure; however, the out‐of‐sample forecasting performance of the two models is approximately equivalent.     

Econometric modelling in the presence of evolutionary change

A methodology is offered which can be used to construct an econometricmodel in the presence of structural change of an evolutionarytype. The theoretical basis for such modelling is drawn fromthe self-organisation approach and operationalised in the contextof the logistic diffusion growth model. The latter is augmentedto allow for the impact of exogenous effects upon both the diffusionrate and boundary limit. We show how the hypothesis of augmentedlogistic diffusion can be falsified using econometric methods.An illustrative case study is used, namely the growth and declineof Australian Building Society Deposits. With the aid of thisexample, it is shown how the approach could be of use to botheconomic forecasters and regulators in conditions of structuralchange where conventional econometric methods are often inappropriate.     

Using Genetic Algorithms to Estimate and Validate Bioeconomic Models: The Case of the Ibero-atlantic Sardine Fishery

Synopsis The Neo-classical approach to fisheries management is based on designing and applying bioeconomic models. Traditionally, the basic bioeconomic models have used pre-established non-linear functional forms (logistic, Cobb–Douglas) in order to try to reflect the dynamics of the renewable resources under study. This assumption might cause misspecification problems and, in consequence, a loss of predictive ability. In this work we intend to verify if there is a bias motivated by employing the said non-linear parametric perspective. For this purpose, we employ a novel non-linear and non-parametric prediction method, called Genetic Algorithms, and we compare its results with those obtained from the traditional methods.     

Cointegration, Efficiency and Forecasting in the Currency Market

Existing literature on using the cointegration approach to examine the efficiency of the foreign exchange market gives mixed results. Arguments typically focus on econometric testing techniques, with fractional cointegration being the most current one. This paper tries to look at the issue from an economic perspective. It shows that the cointegrating relationship, whether cointegrated or fractionally cointegrated, is found mainly among the currencies of the European Monetary System which are set to fluctuate within a given range. Hence, there is no inconsistency with the notion of market efficiency. Yet, exploiting such a cointegrating relationship is helpful in currency forecasting. There is some evidence that restricting the forecasting model to consist of only cointegrated currencies improves forecasting efficiency.     

需求增长型企业销量预测方法--以空调生产企业为例

基于基本时间序列分解法、回归分析法和定性预测法,根据“误差决定权重”的指导原则对各种单一预测方法进行权重分配,提出了适用于需求增长型空调生产企业销量综合性预测方法的基本思路,结合国内某空调生产企业实际调研数据给出了计算实例。     

Accuracy, unbiasedness and efficiency of professional macroeconomic forecasts: An empirical comparison for the G7

In this paper, we use survey data to analyze the accuracy, unbiasedness and efficiency of professional macroeconomic forecasts. We analyze a large panel of individual forecasts that has not previously been analyzed in the literature. We provide evidence on the properties of forecasts for all G7-countries and for four different macroeconomic variables. Our results show a high degree of dispersion of forecast accuracy across forecasters. We also find that there are large differences in the performances of forecasters, not only across countries but also across different macroeconomic variables. In general, the forecasts tend to be biased in situations where the forecasters have to learn about large structural shocks or gradual changes in the trend of a variable. Furthermore, while a sizable fraction of forecasters seem to smooth their GDP forecasts significantly, this does not apply to forecasts made for other macroeconomic variables.     

Advances in forecasting with neural networks? Empirical evidence from the NN3 competition on time series prediction

This paper reports the results of the NN3 competition, which is a replication of the M3 competition with an extension of the competition towards neural network (NN) and computational intelligence (CI) methods, in order to assess what progress has been made in the 10 years since the M3 competition. Two masked subsets of the M3 monthly industry data, containing 111 and 11 empirical time series respectively, were chosen, controlling for multiple data conditions of time series length (short/long), data patterns (seasonal/non-seasonal) and forecasting horizons (short/medium/long). The relative forecasting accuracy was assessed using the metrics from the M3, together with later extensions of scaled measures, and non-parametric statistical tests. The NN3 competition attracted 59 submissions from NN, CI and statistics, making it the largest CI competition on time series data. Its main findings include: (a) only one NN outperformed the damped trend using the sMAPE, but more contenders outperformed the AutomatANN of the M3; (b) ensembles of CI approaches performed very well, better than combinations of statistical methods; (c) a novel, complex statistical method outperformed all statistical and CI benchmarks; and (d) for the most difficult subset of short and seasonal series, a methodology employing echo state neural networks outperformed all others. The NN3 results highlight the ability of NN to handle complex data, including short and seasonal time series, beyond prior expectations, and thus identify multiple avenues for future research.     

Forecasting the term structures of Treasury and corporate yields using dynamic Nelson-Siegel models

We extend Diebold and Li’s dynamic Nelson-Siegel three-factor model to a broader empirical prospective by including the evaluation of the state space approach and by using nine different ratings for corporate bonds. We find that the dynamic Nelson-Siegel factor AR(1) model outperforms other competitors on the out-of-sample forecast accuracy, especially on the investment-grade bonds for the short-term forecast horizon and on the high-yield bonds for the long-term forecast horizon. The dynamic Nelson-Siegel factor state space model, however, becomes appealing on the high-yield bonds in the short-term forecast horizon, where the factor dynamics are more likely time-varying and parameter instability is more probable in the model specification.     

Forecast combinations of computational intelligence and linear models for the NN5 time series forecasting competition

In this work we introduce the forecasting model with which we participated in the NN5 forecasting competition (the forecasting of 111 time series representing daily cash withdrawal amounts at ATM machines). The main idea of this model is to utilize the concept of forecast combination, which has proven to be an effective methodology in the forecasting literature. In the proposed system we attempted to follow a principled approach, and make use of some of the guidelines and concepts that are known in the forecasting literature to lead to superior performance. For example, we considered various previous comparison studies and time series competitions as guidance in determining which individual forecasting models to test (for possible inclusion in the forecast combination system). The final model ended up consisting of neural networks, Gaussian process regression, and linear models, combined by simple average. We also paid extra attention to the seasonality aspect, decomposing the seasonality into weekly (which is the strongest one), day of the month, and month of the year seasonality.