太阳能帆板输出功率短期预测研究

在卫星观测中,经常由于各种原因导致短时间内无法提取目标卫星数据,使控制系统无法继续对其保持高精度跟踪和控制。文章主要研究ARIMA模型在卫星遥测数据短期预测中的应用。使用Matlab软件进行ARIMA模型参数计算,再使用SPSS软件执行数据预测,并进行残差分析,通过实例验证,短时期内的预测数据精度较高。     

Stock-Price Forecasting Based on XGBoost and LSTM

Using time-series data analysis for stock-price forecasting (SPF) is complex and challenging because many factors can influence stock prices (e.g., inflation, seasonality, economic policy, societal behaviors). Such factors can be analyzed over time for SPF. Machine learning and deep learning have been shown to obtain better forecasts of stock prices than traditional approaches. This study, therefore, proposed a method to enhance the performance of an SPF system based on advanced machine learning and deep learning approaches. First, we applied extreme gradient boosting as a feature-selection technique to extract important features from high-dimensional time-series data and remove redundant features. Then, we fed selected features into a deep long short-term memory (LSTM) network to forecast stock prices. The deep LSTM network was used to reflect the temporal nature of the input time series and fully exploit future contextual information. The complex structure enables this network to capture more stochasticity within the stock price. The method does not change when applied to stock data or Forex data. Experimental results based on a Forex dataset covering 2008–2018 showed that our approach outperformed the baseline autoregressive integrated moving average approach with regard to mean absolute error, mean squared error, and root-mean-square error.     

一种混合模型的时序数据异常检测方法

对设备性能指标、用户数据指标的异常检测能有效地发现系统潜在故障,本文提出了一种混合异常检测方法。该方法利用k-means将历史数据按照时间进行划分,使用grubbs算法剔除历史数据中的噪音,并计算各时间段的阈值形成动态阈值,同时利用曲线拟合和ARIMA算法对预处理后的历史数据进行训练,得到对应的模型,作为判断异常的依据。该方法结合了统计学的高效、机器学习的准确,无需对数据进行标注,该方法能自动发现单指标和多指标异常。通过在几个系统的实际运维的检验,本文提出的方法能有效地发现缺数异常和系统异常,提高告警准确率,单指标的查全率达到100%,平均查准率为95.7%,算法的效率满足生产环境中的性能要求。     

Forecasting Method of Stock Market Volatility in Time Series Data Based on Mixed Model of ARIMA and XGBoost

Stock price forecasting is an important issue and interesting topic in financial markets.Because reasonable and accurate forecasts have the potential to generate high economic benefits,many researchers have been involved in the study of stock price forecasts.In this paper,the DWT-ARIMAGSXGB hybrid model is proposed.Firstly,the discrete wavelet transform is used to split the data set into approximation and error parts.Then the ARIMA(0,1,1),ARIMA(1,1,0),ARIMA(2,1,1)and ARIMA(3,1,0)models respectively process approximate partial data and the improved xgboost model(GSXGB)handles error partial data.Finally,the prediction results are combined using wavelet reconstruction.According to the experimental comparison of 10 stock data sets,it is found that the errors of DWT-ARIMA-GSXGB model are less than the four prediction models of ARIMA,XGBoost,GSXGB and DWT-ARIMA-XGBoost.The simulation results show that the DWT-ARIMA-GSXGB stock price prediction model has good approximation ability and generalization ability,and can fit the stock index opening price well.And the proposed model is considered to greatly improve the predictive performance of a single ARIMA model or a single XGBoost model in predicting stock prices.     

混凝土坝变形Wavelet-EGM-PE-ARIMA组合预测模型

混凝土坝的总变形可以归结为由水压和温度变化引起的变形以及随时间发展的变形。其中,水压变形和温度变形体现为总变形中的周期性分量,而时效变形体现为总变形中的趋势性分量。借助复合建模思想,提出一种混凝土坝变形Wavelet-EGM-PE-ARIMA组合预测模型。首先利用小波多分辨分析功能,分解出大坝变形时间序列中的趋势性项、周期性项;其次,运用EGM模型实现对趋势性项的有效预测,采用周期外延模型实现对周期性项的有效预测,在此基础上,利用ARIMA模型实现对EGM模型和周期外延模型残差项的有效预测;最后通过某工程实例,检验所提出模型的有效性。计算结果表明:该组合模型充分考虑大坝各变形分量的变化规律,并基于此,实现对大坝变形时间序列有效的拟合和预测,且其拟合和预测精度均明显优于传统统计模型。     

基于多尺度自回归模型体系的混凝土坝位移预报及工程应用

针对混凝土坝位移监测数据的时频非线性特征严重影响到数值模型预报精度的难题,通过小波技术解析原型数据中多重交叉环境驱动的效应实况,有机结合非线性自回归模型(Nonlinear Autoregressive Model with Exogenous Input, NARX)和差分整合移动平均自回归模型(Autoregressive Integrated Moving Average Model, ARIMA),建立了多尺度组合机制下的自回归模型体系,解决了内蕴复杂混沌特性的监测序列的信息挖掘难点。工程实例分析表明,所建模型的拟合精度及预测能力均得以提升,相比于传统模型具有较好的抗噪性和鲁棒性。此外,所建立的计算模型经一定的优化和拓展,亦可推广应用于其它水工建筑物的效应预报分析。     

基于AR IMA与LOESS的安全生产事故时序预测研究

我国每年安全生产事故都造成大量的人员伤亡和经济损失,因此预防和减少安全生产事故的发生非常重要,利用ARIMA模型和LOESS回归模型组合预测能提高安全生产事故次数预测的精准度.首先建立ARIMA预测模型,用训练集中的预测偏差建立LOESS回归模型,综合两者的预测值,得到最终预测结果.采用2007年9月至2016年7月全国安全生产事故次数数据的实验结果表明:综合2种模型得到的组合预测方法的预测结果精度高于单种模型.     

General deterrence effects of U.S. statutory DUI fine and jail penalties: long-term follow-up in 32 states

INTRODUCTION: We examined effects of state statutory changes in DUI fine or jail penalties for firsttime offenders from 1976 to 2002. METHODS: A quasi-experimental time-series design was used (n=324 monthly observations). Four outcome measures of drivers involved in alcohol-related fatal crashes are: single-vehicle nighttime, low BAC (0.01-0.07g/dl), medium BAC (0.08-0.14g/dl), high BAC (>/=0.15g/dl). All analyses of BAC outcomes included multiple imputation procedures for cases with missing data. Comparison series of non-alcohol-related crashes were included to efficiently control for effects of other factors. Statistical models include state-specific Box-Jenkins ARIMA models, and pooled general linear mixed models. RESULTS: Twenty-six states implemented mandatory minimum fine policies and 18 states implemented mandatory minimum jail penalties. Estimated effects varied widely from state to state. Using variance weighted meta-analysis methods to aggregate results across states, mandatory fine policies are associated with an average reduction in fatal crash involvement by drivers with BAC>/=0.08g/dl of 8% (averaging 13 per state per year). Mandatory minimum jail policies are associated with a decline in single-vehicle nighttime fatal crash involvement of 6% (averaging 5 per state per year), and a decline in low-BAC cases of 9% (averaging 3 per state per year). No significant effects were observed for the other outcome measures. CONCLUSIONS: The overall pattern of results suggests a possible effect of mandatory fine policies in some states, but little effect of mandatory jail policies.     

岷江都江堰降雨与径流变化趋势分析及预测

依据岷江上游紫坪铺水文站监测1955—2008年降雨量与年径流量资料为基础数据,用线性倾向法、滑动平均法、Mann-Kendall法进行系列趋势检验,采用小波分析进行序列周期性分析,用ARIMA模型法构建降雨径流预测模型。结果表明:近54 a来,岷江流域都江堰段降雨量与径流量均有下降趋势,其中降雨下降较快,降雨的突变点比径流少且二者存在6~21 a的周期,得到的ARIMA模型预测精度较高,结合序列趋势变化与周期性,可得到该区未来降雨径流更合理的预测。