基于Box-Jenkins方法的青南高原降水量时间序列分析建模与预测

应用基于Box-Jenkins方法的时间序列分析技术,对青南高原的四个典型地区1961-2005年降水量序列进行ARMA建模分析:验证了四地区年降水量序列的时间序列特性,研究并选择了这些序列的最佳ARMA模型,本文也通过模型对未来降水量进行了预测.模型实证分析的结果表明:在青藏高原降水量时间序列分析建模与预测方面,Box-Jenkins方法及其模型是一种精度较高且切实有效的方法模型.     

ARIMA乘积季节模型及其在传染病发病预测中的应用

本文研究乘积季节模型在传染病发病情况预测中的应用,并探讨提高模型准确性和实用性的途径.以1980年1月至2000年7月江苏省肾综合征出血热发病资料建立模型,以2000年的发病资料作为模型预测效果的考核样本.首先采用差分方法对序列资料进行平稳化,然后进行定阶并估计参数,建立乘积季节模型,最后对预测结果进行检验和分析.从而更好地掌握未来疫情动态发展趋势.检验结果表明,用乘积季节模型对肾综合征出血热月发病情况的拟合结果满意,预测效果良好.     

基于改进的SVR模型在年降水量预测中的应用

鉴于降水量数据的高维非线性性和周期性,建立了支持向量回归(SVR)预测模型用于降水量预测,由于对该模型输入特征的选取极为重要,因此提出了一种基于季节自回归(SARI)的输入特征选取方法.利用已有的降水量数据建立SARI模型,通过观察模型表达式提取建立SVR模型所需的输入特征用于训练支持向量机,并通过网格参数寻优法确定SVR模型的参数,进行降水量预测.实例分析中,应用此模型对黄土丘陵半干旱区域的降水量进行预测,将预测结果与季节时间序列(SARIMA)模型的预测结果进行对比,结果表明,模型具有更高的预测精度和拟合优度,可以用于降水量的预测.     

加权马尔科夫链在鄂尔多斯市年降水量预测中的应用

鄂尔多斯市是内蒙古自治区重要的农牧渔业生产区.本文采用样本均值—标准差分级法,将鄂尔多斯市11个气象站1961-2019年的年降水量序列划分为丰水年、偏丰水年、平水年、偏枯水年,枯水年5个状态;以年降水量序列各阶自相关系数rk为权值,建立了加权马尔科夫预测模型,对鄂尔多斯市2017年、2018年、2019年的年降水量数值和所处状态进行预测.结果显示相对误差分别为1.3%,6.9%,4.5%,预测精度较高,方法得当.于是利用精度检验后的加权马尔科夫模型预测了鄂尔多斯市2020年降水量为312.81mm,划分为平水年.同时利用马尔科夫模型的遍历性,讨论了鄂尔多斯市年降水量的极限分布和丰枯状态重现期.研究结果表明,鄂尔多斯市59a来降水过程中枯水年、偏枯水年、平水年、偏丰水年、丰水年5种状态出现的概率分别为0.1358、0.1941、0.3693、0.1926、0.1083,即出现平水年的可能性最大,重现期为2.71a;出现丰水年的可能性最小,重现期为9.23a.上述研究结果与降水量序列及所处状态是保持一致的,由此可见利用加权马尔科夫链预测鄂尔多斯地区的年降水量是可行且有效的.     

基于模糊时间序列模型的新疆季度降水预测研究

准确预测一个地区未来的降水量,可以提高该地区防灾减灾的能力,更好地为工农业生产生活提供决策参考。本文以新疆1957年至2007年四季降水时间序列为研究对象,基于自动聚类区间划分方法建立模糊时间序列模型,并将该模型应用于新疆2008~2012年的四季降水预报。最后,从拟合精度和预测精度对预测结果进行分析。结果表明,基于模糊时间序列的降水预测模型具有较高预报精度,可用于新疆的季度降水预报。     

加权马尔可夫模型在年降水量上的预测应用

降水是气象预测的重要环节,并在人类的生产生活中扮演着越来越重要的角色.以南京市1971-2011年的年降水量为依据,通过运用加权的马尔可夫模型,分别对其2010、2011年的降水量状态进行预测,并运用向量叠加法来预测出这两年的年降水量具体数值.分析结果表明,该模型在降水量预测方面的应用是可行的,能够有效的提高中长期降水量预测的精度.     

广东省能源消费的ARIMA模型预测分析

对1985-2008年广东省能源消费总量数据序列进行分析,建立了ARIMA(2,2,2)模型,检验结果表明,该模型对原始数据序列有较好的拟合效果,模型的预测效果良好,可用于短期外推预测.     

SARIMA模型与Holt-Winters模型在丙型肝炎发病率预测中的应用

主要探讨SARIMA模型和Holt-Winters模型在丙型肝炎月发病率预测中的应用价值.首先,利用2006年1月至2018年12月的月发病率数据为基础,分别建立SARIMA模型和Holt-Winters模型;然后,预测2019年1月到12月的月发病率数据,并与实际月发病率数据进行比较,得到SARIMA(2,1,1)(0,1,1)₁₂模型的拟合效果较好;最后,利用SARIMA(2,1,1)(0,1,1)₁₂模型预测了2020年1月至12月的月发病率.研究结果将可为相关部门丙型肝炎防控提供科学依据.     

SARIMA模型在新疆手足口病发病率预测中的应用

建立新疆手足口病发病率的季节求和自回归-移动平均模型(Seasonal AutoregressiveIntegrated Moving Average Model,SARIMA),探讨采用SARIMA模型预测手足口病发病趋势的可行性和实用性.利用R统计软件基于新疆2006-2012手足口病月发病率数据建立SARIMA模型,拟合2012年手足口病各月发病率数据,并预测了2013年手足口病月发病率.经过序列平稳化、模型识别以及模型诊断,SARIMA(1,0,1)(0,1,0)_(12)能较好地拟合既往时间段的发病率,且预测值符合新疆手足口病实际发病率的波动趋势.SARIMA模型能够有效地预测手足口病发病趋势,对预警、防控具有积极指导意义.     

基于ARIMA模型的我国大宗农产品价格指数预测

以2006年6月至2015年12月我国大宗农产品价格指数月度时间序列作为研究对象.构建ARIMA(1,1,1)模型对我国大宗农产品价格指数进行了拟合和预测,并对模型拟合效果和预测准确度进行了检验,效果均良好.预测结果表明,从长期变化趋势看,我国大宗农产品价格指数上涨是大势所趋.从短期变化趋势看,大宗农产品面对较大的价格下行压力.     

基于混沌的三江平原月降水时间序列分析

三江平原是我国最大的淡水沼泽区,近年来降水的减少是导致湿地减少的一个重要的自然因素.别拉洪河是三江平原上比较有代表性的沼泽性河流.以别拉洪水文站的降水序列为例,通过相空间重构,分别计算了序列的关联维、最大Lyapunov指数以及Kolmogorov熵等几个序列特征量.计算表明:三江平原月降水序列中是存在明显的混沌特征的,这为以后建立三江平原月降水的混沌预测模型提供了理论依据.     

神经网络汛期降水短期气候预测模型

用1960-2005年每年5-9月安徽宣城7县(市)平均降水量作为预报对象,在对应降水前期逐月74项大气环流特征量资料、500hPa月平均高度场和月平均海温场资料中选取因子.采用主分量分析方法构造网络学习矩阵,降低矩阵维数,提高网络预测模型泛化性能.建立的神经网络汛期降水短期气候预测模型对历年样本拟合精度高,试报效果较好,可在气候预测业务中使用.     

Synoptic airflow and UK daily precipitation extremes

We develop a vector generalised linear model to describe the influence of the atmospheric circulation on extreme daily precipitation across the UK. The atmospheric circulation is represented by three covariates, namely synoptic scale airflow strength, direction and vorticity; the extremes are represented by the monthly maxima of daily precipitation, modelled by the generalised extreme value distribution (GEV). The model parameters for data from 689 rain gauges across the UK are estimated using a maximum likelihood estimator. Within the framework of vector generalised linear models, various plausible models exist to describe the influence of the individual covariates, possible nonlinearities in the covariates and seasonality. We selected the final model based on the Akaike information criterion (AIC), and evaluated the predictive power of individual covariates by means of quantile verification scores and leave-one-out cross validation. The final model conditions the location and scale parameter of the GEV on all three covariates; the shape parameter is modelled as a constant. The relationships between strength and vorticity on the one hand, and the GEV location and scale parameters on the other hand are modelled as natural cubic splines with two degrees of freedom. The influence of direction is parameterised as a sine with amplitude and phase. The final model has a common parameterisation for the whole year. Seasonality is partly captured by the covariates themselves, but mostly by an additional annual cycle that is parameterised as a phase-shifted sine and accounts for physical influences that we have not attempted to explicitly model, such as humidity.     

STOCHASTIC PRICE MODELS AND OPTIMAL TREE CUTTING: RESULTS FOR LOBLOLLY PINE

An empirical investigation of stumpage price models and optimal harvest policies is conducted for loblolly pine plantations in the southeastern United States. The stationarity of monthly and quarterly series of sawtimber prices is analyzed using a unit root test. The statistical evidence supports stationary autoregressive models for the monthly series and for the quarterly series of opening month prices. In contrast, the evidence supports a non-stationary random walk model for the quarterly series of average prices. This conflicting result is likely an artifact of price averaging. The properties of these series significantly affect the forms of optimal price-dependent harvest rules and expected returns. Further, the results have implications for conclusions about market efficiency and the performance of a fixed rotation age.     

Prediction of extremal precipitation by quantile regression forests: from SNU Multiscale Team

This paper considers the problem of spatio-temporal extreme value prediction of precipitation data. This work presents some methods that predict monthly extremes over the next 20 years corresponding to 0.998 quantile at several stations over a certain region. The proposed methods are based on a novel combination of quantile regression forests and circular transformation. As the core of the methodology, quantile regression forests by combining many decorrelated bootstrapping trees may improve prediction performance, and circular transformation is used for building circular transformed predictors of months, which are put into the quantile regression forests model for prediction. The empirical performance of the proposed methods are evaluated through real data analysis, which demonstrates promising results of the proposed approaches.     

A grey modeling procedure based on the data smoothing index for short-term manufacturing demand forecast

Product life cycles have become increasingly shorter because of global competition. Under fierce competition, the use of small samples to establish demand forecasting models is crucial for enterprises. However, limited samples typically cannot provide sufficient information; therefore, this presents a major challenge to managers who must determine demand development trends. To overcome this problem, this paper proposes a modified grey forecasting model, called DSI–GM(1,1). Specifically, we developed a data smoothing index to analyze the data behavior and rewrite the calculation equation of the background value in the applied grey modeling, constructing a suitable model for superior forecasting performance according to data characteristics. Employing a test on monthly demand data of thin film transistor liquid crystal display panels and the monthly average price of aluminum for cash buyers, the proposed modeling procedure resulted in high prediction outcomes; therefore, it is an appropriate tool for forecasting short-term demand with small samples.     

MODELING FOREST PRODUCTIVITY ACROSS LARGE AREAS AND LONG PERIODS

ABSTRACT. There are three classes of forest model used to simulate forest productivity across large areas and over long periods: growth and yield models, based on statistical relationships derived from measurements on trees; the so-called gap models, concerned with species succession and dynamics, and carbon balance or biomass models. The characteristics of each type are discussed and illustrated by reference to some of the more important of the models in current use. The emphasis in this paper is on the carbon balance models, particularly on a new model (3-PG), developed in a deliberate attempt to bridge the gap between growth and yield and carbon balance models, and the companion model (3-PGS) derived from 3-PG to utilize satellite data as inputs to constrain the simulation calculations and improve estimates of growth over time. 3-PG/3-PGS run on monthly time steps, driven by weather data, and avoid the problems of over-parameterization and the requirements for a great deal of input data that limit the practical value of most carbon balance models. We present test results from 3-PG against experimental data, and against forest plot (mensuration) data from large areas; also test results from 3-PGS against estimates of average forest growth over large areas, and in plantations with different planting times, using AVHRR and Landsat MSS data to constrain the model outputs. The paper discusses the problems of the variability of natural forests and the difficulties this causes in validating models intended for use over large areas. The value of remote sensing as means of overcoming this problem is considered.     

Deep learning models for inflation forecasting

We propose a hybrid deep learning model that merges Variational Autoencoders and Convolutional LSTM Networks (VAE-ConvLSTM) to forecast inflation. Using a public macroeconomic database that comprises 134 monthly US time series from January 1978 to December 2019, the proposed model is compared against several popular econometric and machine learning benchmarks, including Ridge regression, LASSO regression, Random Forests, Bayesian methods, VECM, and multilayer perceptron. We find that VAE-ConvLSTM outperforms the competing models in terms of consistency and out-of-sample performance. The robustness of such conclusion is ensured via cross-validation and Monte-Carlo simulations using different training, validation, and test samples. Our results suggest that macroeconomic forecasting could take advantage of deep learning models when tackling nonlinearities and nonstationarity, potentially delivering superior performance in comparison to traditional econometric approaches based on linear, stationary models.     

The Development of a Mathematical Model of the Operations of a National Glass Manufacturer

In this paper, a mathematical model of the entire operations of a national glass manufacturer is developed. This includes the float glass manufacture, distribution, storage operations and the technical considerations dictated by the plant as well as the operating procedures. The model is initially for a planning year, and is generated from a ‘monthly model’. This monthly model interconnects with other monthly models primarily via stock flows. The mathematical model is formulated in a unique way that allows certain production aspects to be modelled using a ‘pseudo-continuous’ time frame, rather than a discrete one. The generation of the overall model (as a mixed integer linear programming problem) and its solution is also discussed.