基于参数模型的智能化预测及其系统研制

研究了基于参数模型的智能化预测、人类专家对预测问题的认知模式与预测模型的智能化选取问题;介绍了所研制的一个基于参数模型的智能化预测系统;应用该系统完成了山东省嘉祥县工商税收的年度预测.     

Combining multiple forecasts given multiple objectives

This paper is concerned with expanding the decision support capabilities of computerized forecasting systems. The expansion allows for the systematic combination of multiple forecasts and the explicit consideration of multiple objectives in the forecast selection process. The methodology used is multiple objective linear programming. Selecting an individual forecast based upon a single objective may not make the best use of available information for a variety of reasons. Combined forecasts may provide a better fit with respect to a single objective than any individual forecast. Even if an individual forecast does provide a good fit with respect to a single objective, a combined forecast may provide a better fit with respect to multiple objectives. An example is used to illustrate the expanded decision support system, its outputs and their properties.     

一种基于需求景气分析和“幼稚预测”思想的需求预测模型

根据汽车生产企业和宾馆企业需求预测问题的基本特点,利用需求时间序列及其变化曲线,提出了一种基于需求景气分析和"幼稚"预测思想的定量预测模型,即首先划分产品的景气变化模式,然后寻找当前点状态的最佳历史匹配,并据此给出需求预测值.本文给出了模型的相关定义及算法,并以宾馆和汽车生产企业的实际需求数据为应用对象,检验了该模型的预测性能.试验结果表明,该方法具有较理想的预测精度.     

From forecasting to ‘la prospective’ a new way of looking at futures

Although‘La Prospective’is not well known in the Anglo-Saxon forecasting literature, it has been for many years widely used in France and other Latin countries with considerable success. Lately, because of the inaccuracy of forecasting and the large forecasting errors that have been experienced, it is suggested that the Prospective approach can be used as a way of dealing with these problems. The main characteristics of‘La Prospective’are that it does not look at the future as a continuation of the past but rather as the outcome of the wishes of various actors and the constraints imposed on them by the environment. Its purpose is to assist in creating alternative futures and then select some alternative that allows for maximum freedom of action.     

Application of support vector machines regression in prediction Shanghai stock composite index

0　IntroductionNeuralnetworks,inrecentyears,havebeensuccessfullyusedformodelingfinancialtimeseries.Researchesrevealneuralnetworkshavebetterresultsindescribingdynamiccharac teristicscomparingwithclassicalstatisticalmethods[1 3] .Back propagation (BP)neuralnetworkisamostpopularlyusedneuralnetworkmodelintheareaoffinancialforecasting .AlthoughBP’sstructureissimple ,ithaspowerfulabilityinsolvingvariousproblems.TherearealsoshortcomingswithBP :needsformanycontrolparameters,difficultyingettingsta…     

The use of monthly indicators to forecast quarterly GDP in the short run: an application to the G7 countries

The delayed release of the National Account data for GDP is an impediment to the early understanding of the economic situation. In the short run, this information gap may be at least partially eliminated by bridge models (BM) which exploit the information content of timely updated monthly indicators. In this paper we examine the forecasting ability of BM for GDP growth in the G7 countries and compare their performance to that of univariate and multivariate statistical benchmark models. We run four alternative one‐quarter‐ahead forecasting experiments to assess BM performance in situations as close as possible to the actual forecasting activity. BM are estimated for GDP both for single countries (USA, Japan, Germany, France, UK, Italy and Canada), and area‐wide (G7, European Union, and Euro area). BM forecasting ability is always superior to that of benchmark models, provided that at least some monthly indicator data are available over the forecasting horizon. Copyright © 2007 John Wiley & Sons, Ltd.     

基于改进多孔算法的时间序列预测

针对小波分析技术存在的边界问题,提出一种改进的多孔算法。使用该算法得到的系数序列,在具备时移不变性的同时,消除了右侧边界存在数据畸变的现象,使小波分析技术结合神经网络等传统预测模型的方法应用于单变量时间序列预测任务具备可行性。为进一步提高预测精度,引入了神经网络集成技术以改善网络泛化能力。实验表明,这种组合预测模型预测效果与稳定性优于传统预测模型。     

基于非线性跟踪—微分器的金融时间序列预测

基于非线性跟踪-微分器的基本原理,开拓性地用二阶离散和三阶离散非线性跟踪-微分器对上证综指进行了预测.从预测结果来看,二阶非线性离散跟踪-微分器预测相对误差控制在8%以内,三阶离散非线性跟踪-微分器预测相对误差控制在5%以内,显示了较好的预测效果.     

基于多输出支持向量回归算法的股市预测

提出了一种新的多输出支持向量回归算法,给出了定义在超球上的损失函数,并将训练SVM转化为迭代解线性方程组,在求解过程中采用边计算边使矩阵降阶的方法,加快了运算速度.建立了该算法应用于股市预测的模型,对上证指数的建模与预测表明:与单输出支持向量回归算法建立的模型相比,该算法具有更好的整体预测精度和抗噪性能,是对股市进行分析和预测的一种可行而有效的方法.