Traditional ARIMA Research Articles

Stock price forecasting based on LLE-BP neural network model

Most of the factors affecting stock prices have data redundancy and nonlinear characteristics. Classical linear mapping dimensional reduction methods such as principal component analysis (PCA) and linear discriminant analysis (LDA) cannot get good results for nonlinear problems. In​ this paper, a local linear embedding dimensional reduction algorithm (LLE) is selected to reduce the dimension of the factors affecting the stock price. The data after dimensional reduction is used as the new input variable of Back Propagation (BP) neural network to realize the stock price prediction. The prediction results are compared with the BP neural network model, PCA-BP model, and the traditional ARIMA (3,1,1) model. The results show that LLE-BP neural network model has higher prediction accuracy in stock price prediction, and it is an effective and feasible stock price prediction method.

Hierarchical time series forecasting via Support Vector Regression in the European Travel Retail Industry

Times series often offers a natural disaggregation in a hierarchical structure. For example, product sales can come from different cities, districts, or states; or be grouped by categories and subcategories. This hierarchical structure can be useful for improving the forecast, and this strategy is known as hierarchical time series (HTS) analysis. In this work, a novel strategy for sales forecasting is proposed using Support Vector Regression (SVR) and hierarchical time series. We formalize three different hierarchical time series approaches: bottom-up SVR, top-down SVR, and middle-out SVR, and use them in a sales forecasting project for the Travel Retail Industry. Various hierarchical structures are proposed for the retail industry in order to achieve accurate product-level predictions. Experiments on these datasets demonstrate the virtues of SVR-based hierarchical time series in terms of predictive performance when compared with the traditional ARIMA and Holt-Winters approaches for this task.

Research and application of the hybrid forecasting model based on secondary denoising and multi-objective optimization for air pollution early warning system

With the increasing irreversible damage caused by air pollution, an early warning system to send warning information to human beings so that they can avoid more harm caused by air pollution is required. A reliable warning system can provide valuable information to protect mankind from the effects of pollution and can act as a tool that allows regulators to implement corresponding measures to reduce air pollution. However, the previous most valuable research studies were focused on pollution forecasting and the extent to which pollution affects health, and the aim of only a few studies was to analyze pollution from an application perspective and to construct a reasonable early warning system. In this study, an air pollution early warning system was constructed, which comprises two modules: an air pollution forecasting module and an air quality evaluation module. In the forecasting module, two denoising methods and a multi-objective optimization algorithm are integrated into a novel hybrid forecasting model. In the evaluation module, fuzzy synthetic evaluation is used to evaluate air quality objectively. To verify the performance of the proposed early warning system, hourly pollutants concentration data were used in a case study of three metropolises in China and three numeric simulation experiments were conducted. The simulation results show that the forecasting performance of the L2,1RF-ELM model used in this study is better than the traditional neural network, and the forecasting model proposed in this paper is better than the traditional statistical model ARIMA. Moreover,the early warning system performed well in terms of highly accurate forecasting and accurate evaluation in the three research areas.

Container flow forecasting through neural networks based on metaheuristics

In this paper we propose a fuzzy neural network prediction approach based on metaheuristics for container flow forecasting. The approach uses fuzzy if–then rules for selection between two different heuristics for developing neural network architecture, simulated annealing and genetic algorithm, respectively. These non-parametric models are compared with traditional parametric ARIMA technique. Time series composed from monthly container traffic observations for Port of Barcelona are used for model developing and testing. Models are compared based on the most important criteria for performance evaluation and for each of the data sets (total container traffic, loaded, unloaded, transit and empty) the appropriate model is selected.

A Deep Learning Integrated Lee–Carter Model

In the field of mortality, the Lee–Carter based approach can be considered the milestone to forecast mortality rates among stochastic models. We could define a “Lee–Carter model family” that embraces all developments of this model, including its first formulation (1992) that remains the benchmark for comparing the performance of future models. In the Lee–Carter model, the κ t parameter, describing the mortality trend over time, plays an important role about the future mortality behavior. The traditional ARIMA process usually used to model κ t shows evident limitations to describe the future mortality shape. Concerning forecasting phase, academics should approach a more plausible way in order to think a nonlinear shape of the projected mortality rates. Therefore, we propose an alternative approach the ARIMA processes based on a deep learning technique. More precisely, in order to catch the pattern of κ t series over time more accurately, we apply a Recurrent Neural Network with a Long Short-Term Memory architecture and integrate the Lee–Carter model to improve its predictive capacity. The proposed approach provides significant performance in terms of predictive accuracy and also allow for avoiding the time-chunks’ a priori selection. Indeed, it is a common practice among academics to delete the time in which the noise is overflowing or the data quality is insufficient. The strength of the Long Short-Term Memory network lies in its ability to treat this noise and adequately reproduce it into the forecasted trend, due to its own architecture enabling to take into account significant long-term patterns.

Application of Temperature Prediction Based on Neural Network in Intrusion Detection of IoT

The security of network information in the Internet of Things faces enormous challenges. The traditional security defense mechanism is passive and certain loopholes. Intrusion detection can carry out network security monitoring and take corresponding measures actively. The neural network-based intrusion detection technology has specific adaptive capabilities, which can adapt to complex network environments and provide high intrusion detection rate. For the sake of solving the problem that the farmland Internet of Things is very vulnerable to invasion, we use a neural network to construct the farmland Internet of Things intrusion detection system to detect anomalous intrusion. In this study, the temperature of the IoT acquisition system is taken as the research object. It has divided which into different time granularities for feature analysis. We provide the detection standard for the data training detection module by comparing the traditional ARIMA and neural network methods. Its results show that the information on the temperature series is abundant. In addition, the neural network can predict the temperature sequence of varying time granularities better and ensure a small prediction error. It provides the testing standard for the construction of an intrusion detection system of the Internet of Things.

Prediction of a New Kind of MR Data

In the field of TD-LTE network problem analysis, compared with traditional methods such as DT(Driver Test) and CQT(Call Quality Test), MR (Measurement Report) has the advantages of comprehensive information and high efficiency, begins to get more and more attention and application. In order to solve the problem of limited open time of measurement data acquisition system, a data prediction method based on LSTM (Long Short-Term Memory) model is proposed. Selecting part of the MR parameters as the experimental object, training LSTM Model with measurement data in a district of Beijing. Experimental results show that the proposed method can predict MR data accurately. Compared with the traditional prediction model ARIMA (Autoregressive Integrated Moving Average Model), this method has lower prediction error and more stable generalization ability.

Characterizing and Modeling Patching Practices of Industrial Control Systems

Industrial Control Systems (ICS) are widely deployed in mission critical infrastructures such as manufacturing, energy, and transportation. The mission critical nature of ICS devices poses important security challenges for ICS vendors and asset owners. In particular, the patching of ICS devices is usually deferred to scheduled production outages so as to prevent potential operational disruption of critical systems. Unfortunately, anecdotal evidence suggests that ICS devices are riddled with security vulnerabilities that are not patched in a timely manner, which leaves them vulnerable to exploitation by hackers, nation states, and hacktivist organizations. In this paper, we present the results from our longitudinal measurement and characterization study of ICS patching behavior. Our study is based on IP scan data collected from Shodan over the duration of three years for more than 500 known industrial ICS protocols and products. Our longitudinal measurements reveal the impact of vulnerability disclosures on ICS patching. Our analysis of more than 100 thousand Internet-exposed ICS devices reveals that about 50% upgrade to newer patched versions within 60 days of a vulnerability disclosure. Based on our measurement and analysis, we further propose a variation of the Bass model to forecast the patching behavior of ICS devices. The evaluation shows that our proposed models have comparable prediction accuracy when contrasted against traditional ARIMA timeseries forecasting models, while requiring less parameters and being amenable to direct physical interpretation.

Prediction of Seawall Settlement Based on a Combined LS-ARIMA Model

The analysis and prediction of seawall settlement are important for seawall engineering maintenance and disaster prevention. Based on the measured seawall settlement time series data, a combined LS-ARIMA forecasting model that fits the trend item by the least-square (LS) method and the season item by the differential self-regression moving average (ARIMA) model was proposed in this study. The monitoring data of one seawall project in Zhejiang, China, is taken as an example to verify the model efficiency and prediction ability. The results show that the prediction accuracy of the new combined LS-ARIMA model was high, with the average relative error (ARE) of 0.23%, much better than that of the traditional ARIMA model (ARE = 0.70%) and the GM (1, 1) model (ARE = 33.43%). This new model has clear physical conception and can effectively improve the prediction accuracy, implying that it can fully tap the dynamic information of monitoring data. The proposed model in this study provides a new research idea for data analysis and prediction of the seawall settlement.

Forecasting zero-inflated price changes with a Markov switching mixture model for autoregressive and heteroscedastic time series

The weekly changes in prices of several German milk-based commodities exhibit not only traditional patterns such as mean dependence and volatility clustering, but also a high frequency of zero changes that cannot be explained by well-known ARIMA-GARCH models. We therefore develop a new mixture model which combines the elements of zero-inflated models that are common in microeconometrics and intermittent demand forecasting with a traditional ARIMA(1,1,0)-GARCH(1,1) model. We describe the model components, the data generation processes, the maximum likelihood estimation techniques, and the generation of forecasting distributions and point forecasts by resampling techniques. The model is applied to a low frequency weekly time series of skimmed whey powder (SWP). Competing submodels are compared using the Akaike information criterion (AIC). Furthermore, in addition to the evaluation of the out-of-sample forecasting performance, several coverage and independence tests are also computed.

Research and Application of a New Hybrid Forecasting Model Based on Genetic Algorithm Optimization: A Case Study of Shandong Wind Farm in China

With the increasing depletion of fossil fuel and serious destruction of environment, wind power, as a kind of clean and renewable resource, is more and more connected to the power system and plays a crucial role in power dispatch of hybrid system. Thus, it is necessary to forecast wind speed accurately for the operation of wind farm in hybrid system. In this paper, we propose a hybrid model called EEMD-GA-FAC/SAC to forecast wind speed. First, the Ensemble empirical mode decomposition (EEMD) can be applied to eliminate the noise of the original data. After data preprocessing, first-order adaptive coefficient forecasting method (FAC) or second-order adaptive coefficient forecasting method (SAC) can be employed to do forecast. It is significant to select optimal parameters for an effective model. Thus, genetic algorithm (GA) is used to determine parameter of the hybrid model. In order to verify the validity of the proposed model, every ten-minute wind speed data from three observation sites in Shandong Peninsula of China and several error evaluation criteria can be collected. Through comparing with traditional BP, ARIMA, FAC, and SAC model, the experimental results show that the proposed hybrid model EEMD-GA-FAC/SAC has the best forecasting performance.

Forecasting Mortality Trends Allowing for Cause-of-Death Mortality Dependence

Longevity risk is among the most important factors to consider for pricing and risk management of longevity products. Past improvements in mortality over many years, and the uncertainty of these improvements, have attracted the attention of experts, both practitioners and academics. Since aggregate mortality rates reflect underlying trends in causes of death, insurers and demographers are increasingly considering cause-of-death data to better understand risks in their mortality assumptions. The relative importance of causes of death has changed over many years. As one cause reduces, others increase or decrease. The dependence between mortality for different causes of death is important when projecting future mortality. However, for scenario analysis based on causes of death, the assumption usually made is that causes of death are independent. Recent models, in the form of Vector Error Correction Models (VECMs), have been developed for multivariate dynamic systems and capture time dependency with common stochastic trends. These models include long-run stationary relations between the variables and thus allow a better understanding of the nature of this dependence. This article applies VECMs to cause-of-death mortality rates to assess the dependence between these competing risks. We analyze the five main causes of death in Switzerland. Our analysis confirms the existence of a long-run stationary relationship between these five causes. This estimated relationship is then used to forecast mortality rates, which are shown to be an improvement over forecasts from more traditional ARIMA processes, which do not allow for cause-of-death dependencies.

Examining deterrence of adult sex crimes: A semi-parametric intervention time-series approach

Motivated by recent developments on dimension reduction (DR) techniques for time-series data, the association of a general deterrent effect with the registration and notification (SORN) policy of South Carolina (SC) for preventing sex crimes was examined. Using adult sex crime arrestee data from 1990 to 2005, the idea of a central mean subspace (CMS) is extended to intervention time-series analysis (CMS-ITS) to model the sequential intervention effects of 1995 (the year SC’s SORN policy was initially implemented) and 1999 (the year the policy was revised to include online notification) on the time-series spectrum. The CMS-ITS model estimation was achieved via kernel smoothing techniques, and compared to interrupted auto-regressive integrated time-series (ARIMA) models. Simulation studies and application to the real data underscore our model’s ability of achieving parsimony, and of detecting intervention effects not earlier determined via traditional ARIMA models. From a public health perspective, findings from this study draw attention to the potential general deterrent effects of SC’s SORN policy. These findings are considered in light of the overall body of research on sex crime arrestee registration and notification policies, which remain controversial.

Forecasting Mortality Trends Allowing for Cause-of-Death Mortality Dependence

Longevity risk is amongst the most important factors to consider for pricing and risk management of longevity products. Past improvements in mortality over many years, and the uncertainty of these improvements, have attracted the attention of experts, both practitioners and academics. Since aggregate mortality rates reflect underlying trends in causes of death, insurers and demographers are increasingly considering cause-of-death data to better understand risks in their mortality assumptions. The relative importance of causes of death has changed over many years. As one cause reduces, others increase or decrease. The dependence between mortality for different causes of death is important when projecting future mortality. However, for scenario analysis based on causes of death, the assumption usually made is that causes of death are independent. Recent models, in the form of Vector Error Correction Models (VECM), have been developed for multivariate dynamic systems and capture time dependency with common stochastic trends. These models include long-run stationary relations between the variables, and thus allow a better understanding of the nature of this dependence. This paper applies VECM to cause-of-death mortality rates in order to assess the dependence between these competing risks. We analyze the five main causes of death in Switzerland. Our analysis confirms the existence of a long-run stationary relationship between these five causes. This estimated relationship is then used to forecast mortality rates, which are shown to be an improvement over forecasts from more traditional ARIMA processes, that do not allow for cause-of-death dependencies.

Wavelet Transform 방법과 SVM 모형을 활용한 상수도 수요량 예측기법 개발

본 연구에서는 Wavelet Transform과 Support Vector Machine (SVM)을 결합한 Hybrid 상수도 수요량 예측 모형을 개발하였다. Wavelet Transform 방법을 활용하여 다양한 스케일이 존재하는 상수도 수요량 시계열을 분해하여 단순한 형태의 시계열로 변환하는데 이용하였으며, 비선형 예측모형인 SVM은 이들 단순화된 시계열을 예측하는데 활용하여 예측성능을 극대화시키는 방안을 수립하였다. 본 연구에서는 상수도 수요량 자료에서 내재되어 있는 주기의 특성과 비선형 예측모형의 장점을 서로 연계한 해석이 가능하였으며 시각적인 검토 및 모든 통계지표에서 개선된 예측결과를 확인할 수 있었다. 특히, 기존 ARIMA 모형 계열에서 나타나는 자기예측문제를 상당부분 개선한 결과를 보여줌으로서 실질적인 수요량 예측모형으로서 활용이 가능할 것으로 판단된다. A hybrid forecasting scheme based on wavelet decomposition coupled to a support vector machine model is presented for water demand series that exhibit nonlinear behavior. The use of wavelet transform followed by the SVM model of each leading component is explored as a model for water demand data. The proposed forecasting model yields better results than a traditional ARIMA time series forecasting model in terms of self-prediction problem as well as reproducing the properties of the observed water demand data by making use of the advantages of wavelet transform and SVM model. The proposed model can be used to substantially and significantly improve the water demand forecasting and utilized in a real operation.

A Zeroinflated ARI(1,1)-IGARCH(1,1) Mixture Model to Forecast Milk-Based Commodity Prices

The weekly change of several German milkbased commodity prices do not only exhibit traditional patterns like mean dependence and volatility clustering, but also a high frequency of zero changes which can not be explained by well known ARIMA-GARCH models. Therefore, we develop a new mixture model which combines the elements of zero-infl ated models common in micro-econometrics and intermittent demand forecasting with a traditional ARIMA(1,1,0)-GARCH(1,1) model. We describe the model components, data generation processes, maximum likelihood estimation techniques and the generation of forecasting distributions and point forecasts by resampling techniques. The model is applied to a low frequency weekly time series of skimmed whey powder (SWP). Competing submodels are selected by likelihood ratio tests and the AIC information criteria. Furthermore, beyond the evaluation of the out-of-sample forecasting performance, several coverage and independence tests are computed.

Modeling Chaotic Behavior of Dhaka Stock Market Index Values Using the Neuro-fuzzy Model

Stock market prediction is an important area of financial forecasting, which attracts great interest of stock investors, stock buyers/sellers, policy makers, applied researchers and many others who are involved in the capital market. This paper aims to develop an efficient model to predict the Dhaka Stock Market Index (DSPI) values using the appropriate forecasting model. It is widely believed that stock data are nonlinear, dynamic and chaotic. In this paper, we propose an adaptive network based fuzzy inference system (ANFIS) to predict DSPI values. We used the daily general DSPI values for the period of March 2003 to October 10, 2006 for the learning and October 11, 2006 to May 31, 2007 for validation. Results obtained by this model are also compared to the back-propagation ANN model and the traditional ARIMA model to show advantages of the proposed ANFIS model. The findings suggest that the ANFIS model can be used as a better predictor for daily general DSPI values as compared to the ANN and the ARIMA models. The review also discussed relevant patents related to our research. Keywords: ANFIS, back-propagation, fuzzy sets and logics, neural network, stock market, time series forecasting, ARCHITECTURES AND LEARNING, ARIMA, DSPI, ANN

Exploiting the Information of Stock Market to Forecast Exchange Rate Movements

The present study examines dynamic relation between stock index and exchange rate by using the daily data for India. The empirical evidence suggests that there is no long-run relationship; however, there is bidirectional causality between stock index and exchange rates. The findings of the causality tests strongly support portfolio or macroeconomic approach on the relationship between exchange rates and stock prices. An attempt is also made to forecast daily returns of INR/USD exchange rates by exploiting the information of causal relationship between exchange rates and stock index using Vector -of-sample performance is benchmarked against the traditional ARIMA model. The potential of the two models is rigorously evaluated by employing a cross-validation scheme and statistical metrics like mean absolute error, root mean square error and directional accuracy. Out-of-sample performance shows that VAR model is robust, and consistently produces superior predictions than ARIMA model.

A hybrid intelligent model for medium-term sales forecasting in fashion retail supply chains using extreme learning machine and harmony search algorithm

A hybrid intelligent (HI) model, comprising a data preprocessing component and a HI forecaster, is developed to tackle the medium-term fashion sales forecasting problem. The HI forecaster firstly adopts a novel learning algorithm-based neural network to generate initial sales forecasts and then uses a heuristic fine-tuning process to obtain more accurate forecasts based on the initial ones. The learning algorithm integrates an improved harmony search algorithm and an extreme learning machine to improve the network generalization performance. Extensive experiments based on real fashion retail data and public benchmark datasets were conducted to evaluate the performance of the proposed model. The experimental results demonstrate that the performance of the proposed model is much superior to traditional ARIMA models and two recently developed neural network models for fashion sales forecasting.

A Hilbert–Huang transform approach for predicting cyber-attacks

A statistical method for prediction and modeling of cyber-attack signal is proposed. The proposed method is developed by coupling the traditional ARIMA method with Hilbert–Huang transform (HHT), designed to reduce the dimensionality and to extract meaningful signals for reliable prediction. HHT decomposes cyber-attack signals of interest into several components including short- and long-term patterns, and random fluctuation. Due to Hilbert transform, the method selects significant decomposed signals that will be employed for signal prediction. Subsequently, by using the traditional dynamic models, the proposed method provides a stable prediction of cyber-attack signal. To show the performance of the proposed method, we analyze daily worm count data from August 1, 2005 to October 9, 2006.