The impact of special days in call arrivals forecasting: A neural network approach to modelling special days

A key challenge for call centres remains the forecasting of high frequency call arrivals collected in hourly or shorter time buckets. In addition to the complex intraday, intraweek and intrayear seasonal cycles, call arrival data typically contain a large number of anomalous days, driven by the occurrence of holidays, special events, promotional activities and system failures. This study evaluates the use of a variety of univariate time series forecasting methods for forecasting intraday call arrivals in the presence of such outliers. Apart from established, statistical methods, we consider artificial neural networks (ANNs). Based on the modelling flexibility of the latter, we introduce and evaluate different methods to encode the outlying periods. Using intraday arrival series from a call centre operated by one of Europe’s leading entertainment companies, we provide new insights on the impact of outliers on the performance of established forecasting methods. Results show that ANNs forecast call centre data accurately, and are capable of modelling complex outliers using relatively simple outlier modelling approaches. We argue that the relative complexity of ANNs over standard statistical models is offset by the simplicity of coding multiple and unknown effects during outlying periods.     

Triple seasonal methods for short-term electricity demand forecasting

Online short-term load forecasting is needed for the real-time scheduling of electricity generation. Univariate methods have been developed that model the intraweek and intraday seasonal cycles in intraday load data. Three such methods, shown to be competitive in recent empirical studies, are double seasonal ARMA, an adaptation of Holt–Winters exponential smoothing for double seasonality, and another, recently proposed, exponential smoothing method. In multiple years of load data, in addition to intraday and intraweek cycles, an intrayear seasonal cycle is also apparent. We extend the three double seasonal methods in order to accommodate the intrayear seasonal cycle. Using six years of British and French data, we show that for prediction up to a day-ahead the triple seasonal methods outperform the double seasonal methods, and also a univariate neural network approach. Further improvement in accuracy is produced by using a combination of the forecasts from two of the triple seasonal methods.     

Detecting intentional herding: what lies beneath intraday data in the Spanish stock market

This paper examines the intentional herd behaviour of market participants, using Li's test to compare the probability distributions of the scaled cross-sectional deviation in returns in the intraday market with the cross-sectional deviation in returns in an ‘artificially created’ market free of intentional herding effects. The analysis is carried out for both the overall market and a sample of the most representative stocks. In addition, a bootstrap procedure is applied in order to gain a deeper understanding of the differences across the distributions under study. The results show that the Spanish market exhibits a significant intraday herding effect that is not detected using other traditional herding measures when familiar and heavily traded stocks are analysed. Furthermore, it is suggested that intentional herding is likely to be better revealed using intraday data, and that the use of a lower frequency data may obscure results revealing imitative behaviour in the market.     

Forecasting time series with multiple seasonal patterns

A new approach is proposed for forecasting a time series with multiple seasonal patterns. A state space model is developed for the series using the innovations approach which enables us to develop explicit models for both additive and multiplicative seasonality. Parameter estimates may be obtained using methods from exponential smoothing. The proposed model is used to examine hourly and daily patterns in hourly data for both utility loads and traffic flows. Our formulation provides a model for several existing seasonal methods and also provides new options, which result in superior forecasting performance over a range of prediction horizons. In particular, seasonal components can be updated more frequently than once during a seasonal cycle. The approach is likely to be useful in a wide range of applications involving both high and low frequency data, and it handles missing values in a straightforward manner.     

Deep learning for energy markets

Deep Learning (DL) is combined with extreme value theory (EVT) to predict peak loads observed in energy grids. Forecasting energy loads and prices is challenging due to sharp peaks and troughs that arise due to supply and demand fluctuations from intraday system constraints. We propose a deep temporal extreme value model to capture these effects, which predicts the tail behavior of load spikes. Deep long‐short‐term memory architectures with rectified linear unit activation functions capture trends and temporal dependencies, while EVT captures highly volatile load spikes above a prespecified threshold. To illustrate our methodology, we develop forecasting models for hourly price and demand from the PJM interconnection. The goal is to show that DL‐EVT outperforms traditional methods, both in‐ and out‐of‐sample, by capturing the observed nonlinearities in prices and demand spikes. Finally, we conclude with directions for future research.     

基于ARMA-GARCH-SN模型的沪深300股指期货日内波动率研究与预测

运用五个交易日的股指期货高频数据(每秒两笔),本文主要研究了沪深300股指期货日内波动率特征并对日内波动率预测。研究发现高频股指期货日内收益率有明显的波动率聚集和条件异方差现象,但无尖峰厚尾现象,收益率序列分布符合有偏正态分布。因此,我们对时间序列建立了最优的ARMA-GARCH-SN模型,并对模型拟合充分性做了验证,拟合结果发现ARMA(1,2)-GARCH(1,1)-SN模型基本能够刻画股指期货高频日内波动特征,条件方差所受的冲击具有很强的持续性、日内波动也具有长记忆性,最后我们还利用自助法对高频股指期货日内波动率两步预测、利用滚动回归预测方法对样本做了样本内预测。预测结果表明,波动率预测结果能够较好地反映股指期货日内波动特征。     

反映强流动曲率效应的非线性湍流模型

首先定性地分析了流线曲率效应对流场湍流结构的影响,然后以U型槽道流为典型算例,对多种湍流模型进行了评估．评估的模型包括:线性涡粘性模型,二阶和三阶非线性涡粘性模型,二阶显式代数应力模型和Reynolds应力模型．评估结果表明,性能良好的三阶非线性涡粘性模型,如黄于宁等人发展的HM模型以及CLS模型,可以较好地描述流线的曲率效应对湍流结构的影响,如凸曲率作用下内壁附近湍流强度的衰减和凹曲率作用下外壁附近湍流的增强,并且较好地确定了管道下游的分离点位置和分离泡长度,其预测的结果和实验符合较好,与Reynolds力模型的结果十分接近,因此可以较好地应用于具有曲率效应的工程湍流的计算．     

基于M-LSTM的股票指数日内交易量分布预测研究

针对现有预测建模方法难以高效提取日内交易量分布复杂变化规律,影响VWAP策略执行效果的问题,本文提出一种MEMD分解下基于LSTM-Attention的股市指数日内交易量分布预测方法M-LSTM。首先,运用MEMD方法将区间多维交易量时序同步分解为若干个独立的本征模态函数(IMF);其次,对各维度分解中高频IMF进行去噪与重构,构建基于LSTM-Attention神经网络的日内交易量分布预测模型,并深入分析股票指数不同走势阶段下模型预测的有效性;最后,分别采用M-LSTM、ARIMA以及SVR等主流方法,对上证指数等四个代表性指数的日内交易量分布进行预测。实验结果表明:M-LSTM预测误差更小,是一种更有效的股票指数日内交易量分布预测方法。     

霍尔特-温特模型在货运量季节性预测中的应用

为了提高货运量季节性预测的精度,应用了一种线性季节预测模型,霍尔特-温特模型。通过傅立叶周期分析确定季节影响和周期长度,利用迭代法寻找模型最优参数。将预测结果与其他三个常用季节预测模型:分组回归模型、可变季节预测模型和季节周期回归模型进行比较,结果表明霍尔特-温特模型预测精度最高。     

基于成交量分解模型的改进VWAP策略

传统VWAP(交易量加权平均价格)策略通过拆分大额委托订单,跟踪市场成交均价,达到最小化冲击成本的目的,而准确预测成交量日内分布是运用VWAP策略的关键。通过详细考察现有的改进VWAP策略中成交量预测模型的建模方式和预测结果,发现由于无法分离成交量日内周期结构,现有模型样本依赖性较大且难以适用于多数股票。因此,本文从个股与市场成交量变化趋势的关系角度出发,推导个股成交量与市场趋势的关系,通过构造个股成交量关于市场因素的因子载荷,将日内成交量分解为市场共同部分和个股特殊部分,预测成交量日内分布并构建动态VWAP策略。实证结果表明新的成交量分解模型可以有效分离个股的成交量日内周期结构,在此基础上构造的改进VWAP策略不仅具有较为广泛的适用性,且跟踪误差减少幅度比现阶段同类型的改进VWAP策略更大,能更好的降低市场冲击成本。     

An improved seasonal rolling grey forecasting model using a cycle truncation accumulated generating operation for traffic flow

Accurate real-time prediction of urban traffic flows is one of the most important problems in traffic management and control optimization research. Short-term traffic flow has complex stochastic and nonlinear characteristics, and it shows a similar seasonality within intraday and weekly trends. Based on these properties, we propose an improved binding cycle truncation accumulated generating operation seasonal grey rolling forecasting model. In the new model, the traffic flow sequence of seasonal fluctuation is converted to a flat sequence using the cycle truncation accumulated generating operation. Then, grey modeling of the cycle truncation accumulated generating operation sequence weakens the stochastic disturbances and highlights the intrinsic grey exponential law after the sequence is accumulated. Finally, rolling forecasts of the limited data reflect the new information priority and timeliness of the grey prediction. Two numerical traffic flow examples from China and Canada, including four groups at different time intervals (1 h, 15 min, 10 min, and 5 min), are used to verify the performance of the new model under different traffic flow conditions. The prediction results show that the model has good adaptability and stability and can effectively predict the seasonal variations in traffic flow. In 15 or 10 min traffic flow forecasts, the proposed model shows better performance than the autoregressive moving average model, wavelet neural network model and seasonal discrete grey forecasting model.     

Combining an Additive and Tree-Based Regression Model Simultaneously: STIMA

Additive models and tree-based regression models are two main classes of statistical models used to predict the scores on a continuous response variable. It is known that additive models become very complex in the presence of higher order interaction effects, whereas some tree-based models, such as CART, have problems capturing linear main effects of continuous predictors. To overcome these drawbacks, the regression trunk model has been proposed: a multiple regression model with main effects and a parsimonious amount of higher order interaction effects. The interaction effects can be represented by a small tree: a regression trunk. This article proposes a new algorithm—Simultaneous Threshold Interaction Modeling Algorithm (STIMA)—to estimate a regression trunk model that is more general and more efficient than the initial one (RTA) and is implemented in the R-package stima. Results from a simulation study show that the performance of STIMA is satisfactory for sample sizes of 200 or higher. For sample sizes of 300 or higher, the 0.50 SE rule is the best pruning rule for a regression trunk in terms of power and Type I error. For sample sizes of 200, the 0.80 SE rule is recommended. Results from a comparative study of eight regression methods applied to ten benchmark datasets suggest that STIMA and GUIDE are the best performers in terms of cross-validated prediction error. STIMA appeared to be the best method for datasets containing many categorical variables. The characteristics of a regression trunk model are illustrated using the Boston house price dataset.

Supplemental materials for this article, including the R-package stima, are available online.     

Transmission dynamics of a switched multi-city model with transport-related infections

Multi-city epidemic models with unrestricted travel, transport-related infection, general nonlinear incidence rate, and seasonality are analyzed. First, a multi-city SIR model is investigated. Seasonality is considered by assuming that the model’s parameters are time-varying and switching. Under this construction, the parameters can be smoothly-varying (for example, due to seasonal changes) or abruptly-varying (for example, due to school holiday breaks). The functional form of the incidence rate is assumed to take a general form that can change in time (for example, due to changes in population behaviour). The effects of transport-related infection and time-varying parameters are studied and some threshold conditions are established which guarantee that the disease-free solution is globally attractive. A screening process and pulse control strategies are applied to the multi-city SIR model in order to investigate and compare the benefits of each strategy. In the pulse control scheme, vaccine failure is considered and the inter-pulse period is not required to equal the seasonal period of the model parameters. Finally, some simulations are given as well as conclusions and future directions.     

Development of improved adaptive approaches to electricity demand forecasting

This paper develops a short-term forecasting system for hourly electricity load demand based on Unobserved Components set up in a State Space framework. The system consists of two options, a univariate model and a non-linear bivariate model that relates demand to temperature. In order to handle the rapidly sampling interval of the data, a multi-rate approach is implemented with models estimated at different frequencies, some of them with ‘periodically amplitude modulated’ properties. The non-linear relation between demand and temperature is identified via a Data-Based Mechanistic approach and finally implemented by Radial Basis Functions. The models also include signal extraction of daily and weekly components. Both models are tested on the basis of a thorough experiment in which other options, like ARIMA and Artificial Neural Networks are also used. The models proposed compare very favourably with the rest of alternatives in forecasting load demand.     

Neural network forecasting for seasonal and trend time series

Neural networks have been widely used as a promising method for time series forecasting. However, limited empirical studies on seasonal time series forecasting with neural networks yield mixed results. While some find that neural networks are able to model seasonality directly and prior deseasonalization is not necessary, others conclude just the opposite. In this paper, we investigate the issue of how to effectively model time series with both seasonal and trend patterns. In particular, we study the effectiveness of data preprocessing, including deseasonalization and detrending, on neural network modeling and forecasting performance. Both simulation and real data are examined and results are compared to those obtained from the Box–Jenkins seasonal autoregressive integrated moving average models. We find that neural networks are not able to capture seasonal or trend variations effectively with the unpreprocessed raw data and either detrending or deseasonalization can dramatically reduce forecasting errors. Moreover, a combined detrending and deseasonalization is found to be the most effective data preprocessing approach.     

基于价格持续时间的中国股市日内风险价值预测

基于高频数据度量日内交易活动的风险是目前日内金融数据与风险管理中极具挑战性的研究课题之一。本文从实时交易的角度,使用中国股市分笔交易数据,基于价格持续时间的自回归条件持续时间(ACD)模型,研究日内不规则交易数据的风险测度,利用日内不等间隔波动模型估计了日内交易的即时条件波动率,对日内不等间隔风险价值进行了预测和检验。实证结果发现日内不等间隔风险价值模型能够比较好的刻画日内交易风险,股票投资者和市场监管者可以基于该工具对日内风险做出合理的预测,达到止损避险和控制风险的目的。     

Model‐based quantification of the volatility of options at transaction level with extended count regression models

In this paper, we elaborate how Poisson regression models of different complexity can be used in order to model absolute transaction price changes of an exchange‐traded security. When combined with an adequate autoregressive conditional duration model, our modelling approach can be used to construct a complete modelling framework for a security's absolute returns at transaction level, and thus for a model‐based quantification of intraday volatility and risk. We apply our approach to absolute price changes of an option on the XETRA DAX index based on quote‐by‐quote data from the EUREX exchange and find that within our Bayesian framework a Poisson generalized linear model (GLM) with a latent AR(1) process in the mean is the best model for our data according to the deviance information criterion (DIC). While, according to our modelling results, the price development of the underlying, the intrinsic value of the option at the time of the trade, the number of new quotations between two price changes, the time between two price changes and the Bid–Ask spread have significant effects on the size of the price changes, this is not the case for the remaining time to maturity of the option. Copyright © 2006 John Wiley & Sons, Ltd.     

Source–sink impulsive bioeconomic models: Seasonal closures with fixed length

The dynamics of four source–sink models for an exploited resource under a constant fishing effort are here presented. Two models are described by ordinary differential equations; the other two are expressed by impulsive differential equations systems. A continuous time growth function for the resource is assumed for each of the four model. The impulsiveness in the harvest activity among fixed seasonal closures were considered in the models expressed by impulsive differential equations. We note that all our models show the possibility of getting a sustainable resource exploitation. The results obtained using both techniques are compared. These metapopulation models suggest the convenience of considering the source patches as marine reserves, in order to preserve the renewable resources.     

Optimal bidding of a virtual power plant on the Spanish day-ahead and intraday market for electricity

We develop a multi-stage stochastic programming approach to optimize the bidding strategy of a virtual power plant (VPP) operating on the Spanish spot market for electricity. The VPP markets electricity produced in the wind parks it manages on the day-ahead market and on six staggered auction-based intraday markets. Uncertainty enters the problem via stochastic electricity prices as well as uncertain wind energy production. We set up the problem of bidding for one day of operation as a Markov decision process (MDP) that is solved using a variant of the stochastic dual dynamic programming algorithm. We conduct an extensive out-of-sample comparison demonstrating that the optimal policy obtained by the stochastic program clearly outperforms deterministic planning, a pure day-ahead strategy, a benchmark that only uses the day-ahead market and the first intraday market, as well as a proprietary stochastic programming approach developed in the industry. Furthermore, we study the effect of risk aversion as modeled by the nested Conditional Value-at-Risk as well as the impact of changes in various problem parameters.