Analysis of pseudo-periodic chronological series with irregularly time-spaced data in view to their prediction. I. Proposed problem

The problem originates from the necessity to predict luminosities of large-amplitude variable stars that are to be observed by the astronomical satellite HIPPARCOS. The data have a specific character: they are unequally time-spaced and can be missing during a long time in comparison to the pseudo-period. So the classical method of time-series analysis must be adapted and new methods are to be searched. In the following papers we present two solutions: a numerical one derived from a Fourier analysis and a symbolic one.     

中国养老保险隐性债务未来规模的预测

伴随转制而产生的隐性债务问题,正在成为制约中国养老保险制度健康发展的"瓶颈".解决养老保险隐性债务的基本前提是要测算债务规模.本文建立养老保险隐性债务的预测模型,对中国养老保险隐性债务规模进行预测并得到结果.     

Selection of the most influential factors on the water-jet assisted underwater laser process by adaptive neuro-fuzzy technique

Water-jet assisted underwater laser cutting has shown some advantages as it produces much less turbulence, gas bubble and aerosols, resulting in a more gentle process. However, this process has relatively low efficiency due to different losses in water. It is important to determine which parameters are the most important for the process. In this investigation was analyzed the water-jet assisted underwater laser cutting parameters forecasting based on the different parameters. The method of ANFIS (adaptive neuro fuzzy inference system) was applied to the data in order to select the most influential factors for water-jet assisted underwater laser cutting parameters forecasting. Three inputs are considered: laser power, cutting speed and water-jet speed. The ANFIS process for variable selection was also implemented in order to detect the predominant factors affecting the forecasting of the water-jet assisted underwater laser cutting parameters. According to the results the combination of laser power cutting speed forms the most influential combination foe the prediction of water-jet assisted underwater laser cutting parameters. The best prediction was observed for the bottom kerf-width (R² = 0.9653). The worst prediction was observed for dross area per unit length (R² = 0.6804). According to the results, a greater improvement in estimation accuracy can be achieved by removing the unnecessary parameter.     

Fuzzy linear regression and its applications to forecasting in uncertain environment

Linear regression has been used for many years in developing mathematical models for application in marketing, management, and sales forecasting. In this paper, two different sales forecasting techniques are discussed. The first technique involves non-fuzzy abstract methods of linear regression and econometrics. A study of the international market sales of cameras, done in 1968 by John Scott Armstrong, utilized these non-fuzzy forecasting techniques. The second sales forecasting technique uses fuzzy linear regression introduced by H. Tanaka, S. Uejima, and K. Asai, in 1980. In this paper, a study of the computer and peripheral equipment sales in the United States is discussed using fuzzy linear regression. Moreover, fuzzy linear regression is applied to forecasting in an uncertain environment. Finally, some possible improvements and suggestions for further study are mentioned.     

How does informational heterogeneity affect the quality of forecasts?

We investigate a toy model of inductive interacting agents aiming to forecast a continuous, exogenous random variable E. Private information on E is spread heterogeneously across agents. Herding turns out to be the preferred forecasting mechanism when heterogeneity is maximal. However in such conditions aggregating information efficiently is hard even in the presence of learning, as the herding ratio rises significantly above the efficient market expectation of 1 and remarkably close to the empirically observed values. We also study how different parameters (interaction range, learning rate, cost of information and score memory) may affect this scenario and improve efficiency in the hard phase.     

Inventory control under temporal demand heteroscedasticity

This paper studies an inventory control problem when the variance of demand is time-varying and exhibits temporal heteroscedasticity. We use a first-order autoregressive process to characterize the dynamic changes in the level of demand over time and a GARCH(1, 1) structure to describe the changes in the variance of demand. Under these demand settings, we quantify the effect of a temporal heterogeneous variance on inventory performance for a system controlled via an order-up-to-level policy. We show that the effect of temporal heteroscedasticity on the forecasting accuracy can be additively decomposed from the total forecasting error variance. The decomposition is used to derive the absolute and relative cost deviations when the temporal heteroscedasticity is ignored. The relationship of these cost deviations to demand autocorrelation and replenishment leadtime is investigated. Computational results show that ignoring temporal heteroscedasticity can increase firm’s inventory costs by as much as 30% when demand autocorrelation is highly positive.     

重庆市浅埋地下洞室安全顶板厚度研究

本文在参考已有资料的基础上, 分析了重庆市浅埋地下洞室安全顶板厚度的影响因素。采用有限单元数值模拟法研究了地基与洞室相互作用下的破坏机制, 以及各影响因素对浅埋地下洞室安全顶板厚度的影响及相关变化规律, 并用逐步线性回归的方法得出了一个能综合考虑各影响因素的安全顶板厚度的预测模型。本文最后还分析了顶拱裂隙对安全顶板厚度的影响。     

Trend mining for system reliability employing a novel heuristic-based Kriging profiling method

This paper presents a novel integration of heuristic-based regressor for the prediction of system reliability. This is implemented by integrating single layer perceptron (SLP) into Kriging model on the basis of an enhanced Particle Swarm Optimization. The proposed method is labeled here as heuristic SLP-based Kriging, or in short HSK. The backbone of HSK is a Competitive Niching-inspired PSO (CNPSO) that serves as the heuristic for identifying the core parameters of the SLP-based Kriging. CNPSO is composed of an opposition-based competitive initialization and a niching-inspired search scheme. For practicality and validation purposes, realistic datasets in the literature of system reliability are considered in the present study. The experimental results obtained demonstrated that HSK outperformed state-of-the-art methods proposed in the literature for addressing the same issue.     

Combining a regression model with a multivariate Markov chain in a forecasting problem

This paper proposes a new concept: the usage of Multivariate Markov Chains (MMC) as covariates. Our approach is based on the observation that we can treat possible categorical (or discrete) regressors, whose values are unknown in the forecast period, as an MMC in order to improve the forecast error of a certain dependent variable. Hence, we take advantage of the information about the past state interactions between the MMC categories to forecast the categorical (or discrete) regressors and improve the forecast of the actual dependent variable.     

Forecasting with cue information: A comparison of multiple regression with alternative forecasting approaches

Multiple linear regression (MLR) is a popular method for producing forecasts when data on relevant independent variables (or cues) is available. The accuracy of the technique in forecasting the impact on Greek TV audience shares of programmes showing sport events is compared with forecasts produced by: (1) a simple bivariate regression model, (2) three different types of artificial neural network, (3) three forms of nearest neighbour analysis and (4) human judgment. MLR was found to perform relatively poorly. The application of Theil’s bias decomposition and a Brunswik lens decomposition suggested that this was because of its inability to handle complex non-linearities in the relationship between the dependent variable and the cues and its tendency to overfit the in-sample data. Much higher accuracy was obtained from forecasts based on a simple bivariate regression model, a simple nearest neighbour procedure and from two of the types of artificial neural network.