Non‐linear,non‐parametric,non‐fundamental exchange rate forecasting

This paper employs a non‐parametric method to forecast high‐frequency Canadian/US dollar exchange rate. The introduction of a microstructure variable, order flow, substantially improves the predictive power of both linear and non‐linear models. The non‐linear models outperform random walk and linear models based on a number of recursive out‐of‐sample forecasts. Two main criteria that are applied to evaluate model performance are root mean squared error (RMSE) and the ability to predict the direction of exchange rate moves. The artificial neural network (ANN) model is consistently better in RMSE to random walk and linear models for the various out‐of‐sample set sizes. Moreover, ANN performs better than other models in terms of percentage of correctly predicted exchange rate changes. The empirical results suggest that optimal ANN architecture is superior to random walk and any linear competing model for high‐frequency exchange rate forecasting. Copyright © 2006 John Wiley & Sons, Ltd.     

Changing time scale for short-term forecasting in financial markets

A forecasting model based on high-frequency market makers quotes of financial instruments is presented. The statistical behaviour of these time series leads to discussion of the appropriate time scale for forecasting. We introduce variable time scales in a general way and define the new concept of intrinsic time. The latter reflects better the actual trading activity. Changing time scale means forecasting in two steps, first an intrinsic time forecast against physical time, then a price forecast against intrinsic time. The forecasting model consists, for both steps, of a linear combination of non-linear price-based indicators. The indicator weights are continuously re-optimized through a modified linear regression on a moving sample of past prices. The out-of-sample performance of this algorithm is reported on a set of important FX rates and interest rates over many years. It is remarkably consistent. Results for short horizons as well as techniques to measure this performance are discussed.     

On long memory origins and forecast horizons

Most long memory forecasting studies assume that long memory is generated by the fractional difference operator. We argue that the most cited theoretical arguments for the presence of long memory do not imply the fractional difference operator and assess the performance of the autoregressive fractionally integrated moving average (ARFIMA) model when forecasting series with long memory generated by nonfractional models. We find that ARFIMA models dominate in forecast performance regardless of the long memory generating mechanism and forecast horizon. Nonetheless, forecasting uncertainty at the shortest forecast horizon could make short memory models provide suitable forecast performance, particularly for smaller degrees of memory. Additionally, we analyze the forecasting performance of the heterogeneous autoregressive (HAR) model, which imposes restrictions on high-order AR models. We find that the structure imposed by the HAR model produces better short and medium horizon forecasts than unconstrained AR models of the same order. Our results have implications for, among others, climate econometrics and financial econometrics models dealing with long memory series at different forecast horizons.     

A comparison of conditional predictive ability of implied volatility and realized measures in forecasting volatility

In a conditional predictive ability test framework, we investigate whether market factors influence the relative conditional predictive ability of realized measures (RMs) and implied volatility (IV), which is able to examine the asynchronism in their forecasting accuracy, and further analyze their unconditional forecasting performance for volatility forecast. Our results show that the asynchronism can be detected significantly and is strongly related to certain market factors, and the comparison between RMs and IV on average forecast performance is more efficient than previous studies. Finally, we use the factors to extend the empirical similarity (ES) approach for combination of forecasts derived from RMs and IV.     

A new Bayesian formulation for Holt's exponential smoothing

In this paper we propose a Bayesian forecasting approach for Holt's additive exponential smoothing method. Starting from the state space formulation, a formula for the forecast is derived and reduced to a two‐dimensional integration that can be computed numerically in a straightforward way. In contrast to much of the work for exponential smoothing, this method produces the forecast density and, in addition, it considers the initial level and initial trend as part of the parameters to be evaluated. Another contribution of this paper is that we have derived a way to reduce the computation of the maximum likelihood parameter estimation procedure to that of evaluating a two‐dimensional grid, rather than applying a five‐variable optimization procedure. Simulation experiments confirm that both proposed methods give favorable performance compared to other approaches. Copyright © 2008 John Wiley & Sons, Ltd.     

A diffusion model for products with indirect network externalities

This paper develops a new diffusion model that incorporates the indirect network externality. The market with indirect network externalities is characterized by two‐way interactive effects between hardware and software products on their demands. Our model incorporates two‐way interactions in forecasting the diffusion of hardware products based on a simple but realistic assumption. The new model is parsimonious, easy to estimate, and does not require more data points than the Bass diffusion model. The new diffusion model was applied to forecast sales of DVD players in the United States and in South Korea, and to the sales of Digital TV sets in Australia. When compared to the Bass and NSRL diffusion models, the new model showed better performance in forecasting long‐term sales. Copyright © 2008 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.     

Judgement in learning‐curve forecasting: a laboratory study

This study investigates whether human judgement can be of value to users of industrial learning curves, either alone or in conjunction with statistical models. In a laboratory setting, it compares the forecast accuracy of a statistical model and judgemental forecasts, contingent on three factors: the amount of data available prior to forecasting, the forecasting horizon, and the availability of a decision aid (projections from a fitted learning curve). The results indicate that human judgement was better than the curve forecasts overall. Despite their lack of field experience with learning curve use, 52 of the 79 subjects outperformed the curve on the set of 120 forecasts, based on mean absolute percentage error. Human performance was statistically superior to the model when few data points were available and when forecasting further into the future. These results indicate substantial potential for human judgement to improve predictive accuracy in the industrial learning‐curve context. Copyright © 1999 John Wiley & Sons, Ltd.     

Forecasting carbon price using a multi-objective least squares support vector machine with mixture kernels

For improving forecasting accuracy and trading performance, this paper proposes a new multi-objective least squares support vector machine with mixture kernels to forecast asset prices. First, a mixture kernel function is introduced into taking full use of global and local kernel functions, which is adaptively determined following a data-driven procedure. Second, a multi-objective fitness function is proposed by incorporating level forecasting and trading performance, and particle swarm optimization is used to synchronously search the optimal model selections of least squares support vector machine with mixture kernels. Taking CO₂ assets as examples, the results obtained show that compared with the popular models, the proposed model can achieve higher forecasting accuracy and higher trading performance. The advantages of the mixture kernel function and the multi-objective fitness function can improve the forecasting ability of the asset price. The findings also show that the models with a high-level forecasting accuracy cannot always have a high trading performance of asset price forecasting. In contrast, high directional forecasting usually means a high trading performance.     

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.     

An exploratory analysis of portfolio managers' probabilistic forecasts of stock prices

This study reports the results of an experiment that examines (1) the effects of forecast horizon on the performance of probability forecasters, and (2) the alleged existence of an inverse expertise effect, i.e., an inverse relationship between expertise and probabilistic forecasting performance. Portfolio managers are used as forecasters with substantive expertise. Performance of this ‘expert’ group is compared to the performance of a ‘semi-expert’ group composed of other banking professionals trained in portfolio management. It is found that while both groups attain their best discrimination performances in the four-week forecast horizon, they show their worst calibration and skill performances in the 12-week forecast horizon. Also, while experts perform better in all performance measures for the one-week horizon, semi-experts achieve better calibration for the four-week horizon. It is concluded that these results may signal the existence of an inverse expertise effect that is contingent on the selected forecast horizon.     

An outlier robust hierarchical Bayes model for forecasting: the case of Hong Kong

This paper introduces a Bayesian forecasting model that accommodates innovative outliers. The hierarchical specification of prior distributions allows an identification of observations contaminated by these outliers and endogenously determines the hyperparameters of the Minnesota prior. Estimation and prediction are performed using Markov chain Monte Carlo (MCMC) methods. The model forecasts the Hong Kong economy more accurately than the standard V AR and performs in line with other complicated BV AR models. It is also shown that the model is capable of finding most of the outliers in various simulation experiments. Copyright © 2004 John Wiley & Sons, Ltd.     

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.     

改进果蝇优化算法优化广义回归神经网络的预测问题研究

为提高传统非线性预测模型的预测精度,提出一种基于改进果蝇优化算法优化广义回归神经网络的预测方法,将果蝇群体分两部分分别进行迭代寻优,从而改进了果蝇优化算法的寻优性能,进而避免了在寻优过程中陷入局部最优。该方法利用改进果蝇优化算法优化广义回归神经网络的径向基函数扩展参数,然后用训练好的广义回归神经网络预测模型进行预测,最后通过订单预测算例进行实证研究。实证研究结果显示,该方法在解决订单预测问题中与未改进的果蝇优化算法优化广义回归神经网络和传统的广义回归神经网络方法对比,具有更高的预测精度和更好的非线性拟合能力。     

Forecasting air pollution PM2.5 in Beijing using weather data and multiple kernel learning

PM_2.5 mass concentration prediction is an important research issue because of the increasing impact of air pollution on the urban environment. In this paper, a PM_2.5 forecasting framework incorporating meteorological factors based on multiple kernel learning (MKL) is proposed to forecast the near future PM_2.5. In addition, we develop a novel two-step algorithm for solving the primal MKL problem. Compared with most existing MKL 2-step algorithms, the proposed algorithm does not require the optimal step size for updating kernel combination coefficients by linear search. To demonstrate the performance of the proposed forecasting framework, its performance is compared to single kernel-based support vector regression (SVR). Data sets of an inland city Beijing acquired from UCI are used to train and validate both of two methods. Experiments show that our proposed method outperforms the SVR.     

Predictive likelihood for coherent forecasting of count time series

A new forecasting method based on the concept of the profile predictive likelihood function is proposed for discrete‐valued processes. In particular, generalized autoregressive moving average (GARMA) models for Poisson distributed data are explored in detail. Highest density regions are used to construct forecasting regions. The proposed forecast estimates and regions are coherent. Large‐sample results are derived for the forecasting distribution. Numerical studies using simulations and two real data sets are used to establish the performance of the proposed forecasting method. Robustness of the proposed method to possible misspecifications in the model is also studied.     

Forecasting using high‐frequency data: a comparison of asymmetric financial duration models

The first purpose of this paper is to assess the short‐run forecasting capabilities of two competing financial duration models. The forecast performance of the Autoregressive Conditional Multinomial–Autoregressive Conditional Duration (ACM‐ACD) model is better than the Asymmetric Autoregressive Conditional Duration (AACD) model. However, the ACM‐ACD model is more complex in terms of the computational setting and is more sensitive to starting values. The second purpose is to examine the effects of market microstructure on the forecasting performance of the two models. The results indicate that the forecast performance of the models generally decreases as the liquidity of the stock increases, with the exception of the most liquid stocks. Furthermore, a simple filter of the raw data improves the performance of both models. Finally, the results suggest that both models capture the characteristics of the micro data very well with a minimum sample length of 20 days. Copyright © 2008 John Wiley & Sons, Ltd.     

Linear combination of restrictions and forecasts in time series analysis

An important tool in time series analysis is that of combining information in an optimal way. Here we establish a basic combining rule of linear predictors and show that such problems as forecast updating, missing value estimation, restricted forecasting with binding constraints, analysis of outliers and temporal disaggregation can be viewed as problems of optimal linear combination of restrictions and forecasts. A compatibility test statistic is also provided as a companion tool to check that the linear restrictions are compatible with the forecasts generated from the historical data. Copyright © 2000 John Wiley & Sons, Ltd.     

Forecasting contemporaneous aggregates and the combination of forecasts: The case of the U.K. monetary aggregates

A number of papers in recent years have investigated the problems of forecasting contemporaneously aggregated time series and of combining alternative forecasts of a time series. This paper considers the integration of both approaches within the example of assessing the forecasting performance of models for two of the U.K. monetary aggregates, £M3 and MO. It is found that forecasts from a time series model for aggregate £M3 are superior to aggregated forecasts from individual models fitted to either the components or counterparts of £M3 and that an even better forecast is obtained by forming a linear combination of the three alternatives. For MO, however, aggregated forecasts from its components prove superior to either the forecast from the aggregate itself or from a linear combination of the two.     

Macroeconomic forecasting with mixed data sampling frequencies: Evidence from a small open economy

The aim of this study was to forecast the Singapore gross domestic product (GDP) growth rate by employing the mixed‐data sampling (MIDAS) approach using mixed and high‐frequency financial market data from Singapore, and to examine whether the high‐frequency financial variables could better predict the macroeconomic variables. We adopt different time‐aggregating methods to handle the high‐frequency data in order to match the sampling rate of lower‐frequency data in our regression models. Our results showed that MIDAS regression using high‐frequency stock return data produced a better forecast of GDP growth rate than the other models, and the best forecasting performance was achieved by using weekly stock returns. The forecasting result was further improved by performing intra‐period forecasting.