ARMA时间序列模型的研究与应用

ARMA模型是一种最常见的重要的时间序列模型,它被广泛应用到各种行业预测中,本文在给出ARMA三种模式和实现方法的同时,给出ARMA模型在股市应用的一个实例。     

基于ARMA模型的电视台收视率预测方法设计和实现

收视率是电视行业中最重要的指标之一,目前行业内预测收视率的方法均没有考虑收视率自身的长期发展趋势,忽略了预测过程中重要的记录信息,导致预测结果精度不够,充分考虑收视率数据自身长期稳定性的特点,使用ARMA模型来进行预测,使预测精度达到用户期望。ARMA模型是一类常用的随机时序模型,它是一种精度较高的时序预测方法,预测过程中充分考虑序列内在的发展趋势,对于具有长期稳定性的时间序列的预测非常有效,并已经在各个行业得到了应用,预测精度较高。收视率属于长期稳定,起伏波动不大的时间序列,引入ARMA模型进行预测必将改善传统预测方法得到结果的精度,仿真结果表明使用ARMA模型可以提高收视率预测精度。     

基于ARMA和BP_AdaBoost的组合销售预测模型研究

为了提高销售预测的准确性,建立了组合销售预测模型。历史销售数据是非线性、时变的时间序列,可看成由线性和非线性2部分组成。用ARMA模型预测线性部分,用BP_Ada Boost模型预测非线性部分,然后将2部分预测结果叠加得到销售预测结果。该组合模型克服了单纯采用ARMA模型预测结果精度低的问题,也克服了单纯使用BP神经网络模型容易陷入局部极小值的问题。经实验对比表明,采用组合预测模型能够更加准确、全面地反应销售规律,提高了销售预测的准确性。     

组合ARMA与SVR模型的时间序列预测

经典的ARMA模型常用于平稳时间序列的预测,而对于自然界绝大部分的非平稳序列一般采用确定性时序分析和随机时序分析.确定性时序分析对随机性信息浪费严重,而随机时序分析经过差分平稳序列后又回归到ARMA模型.本文利用在充分ARMA模型拟合后的残差序列进行支持向量回归(SVR)拟合,进而对原序列进行组合预测,比起单一模型的拟合及预测,该组合有效地提高了预测精度.     

时序算法在销售预测中的应用研究

在金融市场、信息网络以及电子商务等领域中积累了大量时间序列数据,对这些数据进行深层次的分析,是数据挖掘研究中的重要方向之一.Microsoft时序算法是一个新的预测算法,使用Microsoft时序算法创建的模型是一个自动回归模型.本文用Micmsoft时序算法建立预测模型,预测某食品零售超市未来的销售情况.     

时间序列分析方法及ARMA,GARCH两种常用模型

证券市场具有数据单一性（大量不需要经过特殊处理的数据）、分析手段多样性和隐蔽性的特点,且与其飞速发展不相称的是证券分析技术进展的缓慢。股市系统中时间序列的预测问题具有重要的理论及实际意义,时间序列的获取是通过对数据库中数据进行分类汇总分析而获得。获取时间序列数据以后可以对它进行预测分析,从而较准确地预见系统的演进。文中介绍了时间序列的基本知识,同时比较了ARMA和GARCH两种常用模型,得出对于中国股市,GARCH模型性能优于ARCH模型。     

时间序列分析方法及ARMA，GARCH两种常用模型

证券市场具有数据单一性（大量不需要经过特殊处理的数据）、分析手段多样性和隐蔽性的特点,且与其飞速发展不相称的是证券分析技术进展的缓慢。股市系统中时间序列的预测问题具有重要的理论及实际意义,时间序列的获取是通过对数据库中数据进行分类汇总分析而获得。获取时间序列数据以后可以对它进行预测分析,从而较准确地预见系统的演进。文中介绍了时间序列的基本知识,同时比较了ARMA和GARCH两种常用模型,得出对于中国股市,GARCH模型性能优于ARCH模型。     

基于小波分解和ARMA模型的沪深300股指期货基差预测

一般说来,若时间序列满足平稳随机过程的性质,则可用经典的ARMA模型进行建模和预测,这是常用的研究思路和手段.但是,金融时间序列由于随机波动较大,很少满足经典ARMA模型的适用条件,无法直接采用该模型进行处理.通过小波分析处理后,将原本非平稳的序列处理为近似平稳的序列,可以采用ARMA模型进行建模和分析.在本文中,采用此思路对沪深300股指期货日内高频数据进行分析,表明该方法对金融数据建模的有效性.     

基于ARMA模型的煤矿瓦斯灾害预测研究

瓦斯预测有助于减少煤矿瓦斯灾害损失,甚至可以完全避免严重事故或灾害的发生.使用时间序列分析法,建立瓦斯灾害预测的自回归滑动平均ARMA模型,用AIC信息量准则实现模型定阶,用最小二乘法确定模型中的未知参数.对于非平稳时间序列,经差分处理后得到平稳时间序列,再用ARMA模型进行预测.仿真结果表明建立的预测模型和数据处理方法能获得较准确的预测结果.     

一种新的统计预测模型——多项式系数自回归模型

传统的自回归滑动平均模型（ARMA）和新近出现的函数系数自回归模型（FAR）不能满足非线性时间序列预测分析的准确度与运算速度要求,为了改进预测性能,研究提出了一种新的统计预测模型——多项式系数自回归模型（PCAR）。给出了PCAR模型的表示形式,详细探讨了PCAR模型的参数估计和阶次选择方法,在此基础上又提出了基于BIC准则的建模算法。同ARMA模型相比,PCAR模型扩大了适用对象范围,有效降低了模型选择误差;同FAR模型相比,它具有参数模型的特点,避免了系数函数局部线性回归估计所存在的不足;分析了PCAR模型与ARMA、FAR模型的等价条件。通过实验分析得出了PCAR模型较ARMA、FAR模型的单步预测准确度分别提高了99.65%和18.7%的结论,而且PCAR建模运算所需时间仅为FAR模型的0.2%。     

基于聚类和ARMA时间序列的I/O区域预取

预取是提高存储系统性能的主要手段之一.但现有存储系统的设备层并不知道任何I/O访问的语义信息,因而不能充分利用I/O访问的语义来预取下一时刻要访问的数据,只能利用较简单的方式如I/O访问的局部性、顺序访问和循环访问等特性来实现简单的预测.为此,本文根据存储系统的特点提出了实用且高效的基于连续度的聚类算法来发现密集读请求访问的区域,并采用ARMA时间序列模型来预测密集读请求可能访问的区域及访问时刻,为正确的预取提供了准确的信息.为提高预取的准确性,并采用了动态参数估计的策略.通过大量实验的结果验证了这两种算法的正确性和预测的准确性,能较大的提高存储系统的预取效率.     

基于连续度聚类和动态ARMA时间序列预测I／O区域

为了动态优化存储系统中数据的分布，存储系统需要能动态发现密集I／O区域和预测未来密集I／O访问的区域，并根据发现和预测的结果来指导存储系统的优化。为此，本文根据存储系统的特点提出了实用且高效的基于连续度的聚类算法来发现密集I／O访问的区域，并采用ARMA时间序列模型来预测密集I／O可能访问的区域。为提高预测的准确性，采用了动态参数估计的策略。通过大量实验的结果验证了这两种算法的正确性和预测的准确性，对存储系统的优化具有较好的指导作用。     

Data-driven models for monthly streamflow time series prediction

Data-driven techniques such as Auto-Regressive Moving Average (ARMA), K-Nearest-Neighbors (KNN), and Artificial Neural Networks (ANN), are widely applied to hydrologic time series prediction. This paper investigates different data-driven models to determine the optimal approach of predicting monthly streamflow time series. Four sets of data from different locations of People’s Republic of China (Xiangjiaba, Cuntan, Manwan, and Danjiangkou) are applied for the investigation process. Correlation integral and False Nearest Neighbors (FNN) are first employed for Phase Space Reconstruction (PSR). Four models, ARMA, ANN, KNN, and Phase Space Reconstruction-based Artificial Neural Networks (ANN-PSR) are then compared by one-month-ahead forecast using Cuntan and Danjiangkou data. The KNN model performs the best among the four models, but only exhibits weak superiority to ARMA. Further analysis demonstrates that a low correlation between model inputs and outputs could be the main reason to restrict the power of ANN. A Moving Average Artificial Neural Networks (MA-ANN), using the moving average of streamflow series as inputs, is also proposed in this study. The results show that the MA-ANN has a significant improvement on the forecast accuracy compared with the original four models. This is mainly due to the improvement of correlation between inputs and outputs depending on the moving average operation. The optimal memory lengths of the moving average were three and six for Cuntan and Danjiangkou, respectively, when the optimal model inputs are recognized as the previous twelve months.     

ARMA(p,q) type high order fuzzy time series forecast method based on fuzzy logic relations

Within classic time series approaches, a time series model can be studied under 3 groups, namely AR (autoregressive model), MA (moving averages model) and ARMA (autoregressive moving averages model). On the other hand, solutions are based mostly on fuzzy AR time series models in the fuzzy time series literature. However, just a few fuzzy ARMA time series models have proposed until now. Fuzzy AR time series models have been divided into two groups named first order and high order models in the literature, highlighting the impact of model degree on forecast performance. However, model structure has been disregarded in these fuzzy AR models. Therefore, it is necessary to eliminate the model specification error arising from not utilizing of MA variables in the fuzzy time series approaches. For this reason, a new high order fuzzy ARMA(p,q) time series solution algorithm based on fuzzy logic group relations including fuzzy MA variables along with fuzzy AR variables has been proposed in this study. The main purpose of this article is to show that the forecast performance can be significantly improved when the deficiency of not utilizing MA variables. The other aim is also to show that the proposed method is better than the other fuzzy ARMA time series models in the literature from the point of forecast performance. Therefore, the new proposed method has been compared regarding forecast performance against some methods commonly used in literature by applying them on gold prices in Turkey, Istanbul Stock Exchange (IMKB) and the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX).     

基于小波的非平稳时间序列预测方法研究

基于小波分析技术,将原始非平稳时间序列分解为一层近似系数和多层细节系数,对其分别采用自回归滑动平均模型以及BP神经网络模型,对各层系数进行建模与预测;通过整合各层系数,得到原始时间序列的预测值。运用这种方法对因特网某节点网络流量数据和某地区日最高气温数据进行预测的结果表明,建立在小波分解基础上的这两种方法都能够有效地应用于非平稳时间序列的预测;而小波-BP神经网络的预测方法无论是精度还是计算复杂度方面都要明显优于小波-ARMA方法。     

非线性时间序列预报的隐多分辨ARMA模型

研究一类用于非线性时间序列预报的隐多分辨自回归滑动平均(ARMA)模型,该模型以ARMA模型为初始细水平模型(即隐多分辨模型的基本块)．证明了模型的建模精度由水平问的方差决定．研究了新模型的自相关函数结构,给出了参数估计的Bayes方法和Metropolis-Hasting算法．进一步提出了一种可以直接用于不同基本块的隐多分辨模型的非线性时间序列预报方法,证明了其比其他的线性预报方法和隐多分辨模型预报方法降低了预报误差．最后通过数值模拟和实例验证了模型和预报方法,并和其他模型进行比较,结果表明新提出模型和预报方法能够更好地描述数据的特征,提高预报的精度．     

小波变换与GARCH组合模型的网络流量预测

在一些网络环境当中,网络流量具有非线性、异方差性和波动集群现象,传统的小波变换与ARMA组合模型不能很好地描述网络流量的这些特性。因此,研究使用了小波变换与广义自回归条件异方差GARCH组合模型来预测网络流量。首先,使用小波变换原理将网络流量序列分解成高频部分和低频部分,在此基础上对各个子序列分别建立相应的GARCH模型并进行预测;然后,使用小波变换原理将各个子序列的预测结果进行重构,从而最终实现对原始网络流量的预测。通过仿真实验表明,该模型的预测精度较之传统的小波变换与ARMA组合模型的预测精度得到了大幅提升。     

An integrated framework of genetic network programming and multi-layer perceptron neural network for prediction of daily stock return: An application in Tehran stock exchange market

Evolutionary algorithms are generally used to find or generate the best individuals in a population. Whenever these algorithms are applied to agent systems, they will lead to optimal solutions. Genetic Network Programming (GNP), which contains graph networks, is one of the developed evolutionary algorithms. When the aim is to forecast the share price or return, ascending and descending trends, volatilities, recent returns, fundamental and technical factors have remarkable impacts on the prediction. This is why technical indicators are used to constitute a set of trading rules. In this paper, we apply an integrated framework consisting of GNP model along with a reinforcement learning and Multi-Layer Perceptron (MLP) neural network to classify data and also time series models to forecast the stock return. Moreover, we utilize rules of accumulation based on the GNP model’s results to forecast the return. The aim of using these models alongside one another is to estimate one-day return. The results derived from 9 stocks with regard to the Tehran Stock Exchange Market. GNP extracts a prodigious number of rules on the basis of 5 technical indicators with 3 times period. Next, MLP network classifies data and finds the similarity between future data and past data concerning a stock (5 sub-period) through classification. Subsequently, a number of conditions are established, in order to choose the best estimation between GNP-RL and ARMA. Distinct comparison with the ARMA–GARCH model, which is operated for return estimation and risk measurement in many researches, demonstrates an extended forecasting power of the proposed model, by the name of GNP–ARMA, reducing error by a mean of 16%.