The bidirectional reflectance distribution function (BRDF) is simplefied and modelled according to the principles of simple geometry optics and statistical modeling.

Aiming at the issues that the soft-sensor models have only narrow working scopes and are difficult to reflect material property changes in the petroleum industry product quality the soft-sensor method is discussed based on physical analysis and statistics modeling.

Seemingly unrelated regression model are multiple regression equation systems which can be abbreviate as SUR. It differs from the multivariate regressions model in that it allows difference explanatory variables between difference equations. This feature let great flexibility to statistics modeling.

似乎不相关回归模型(seemingly unrelated regressions model SUR)是由多个回归方程组成的方程组,它与多元回归模型(multivariate regression model)的区别在于允许各方程存在不同的自变量,这样的特性给统计建模带来很大的灵活性。

The algorithm structure, implement procedure and application field etc of the temporal difference algorithms including the adjacent frame difference mean algorithm and three frame difference algorithm and the background statistics modeling algorithms including background subtraction and Gaussian Mixture Model are particularly discussed in this paper.

By comparing the errors-of-fit produced by both the new model and the conventional target models, we conclude that the new nonparametric method of statistic modeling is superior, for it makes the statistic modeling of radar target easier and more exact.

This paper based on the present status of domestic and outside equipment and productionof rubber industry, considered the process of rubber mixing technology, used statistic modeling, NN andexpert system and advanced calculation technology to realize the intelligent control of rubber mixing.

Hidden Markov Models (HMM) is a kind of statistic modeling tool been used successfully in speech recognition. The paper brings the technology of HMM into the study of filtering noise in digital discrete signal first and brings up HMM averaging method. It proves efficient to use HMM in the work after computer rebuilding simulations of noise-adding harmonic wave' rectangular wave.

Based on the strong relationship of the three variables (speed, direction and height) determining aircraft moving conditions, This paper deals with the statistic modeling of aircraft perturbation motion in the method of multidimensional self-recurrence in time sequences and proves through instances that the method presented is feasible and effective.

The results of analytical and statistical modeling of the stability of the amplitude-frequency characteristics are presented.

The problem of strong evaporation of matter filling periodic rectangular semi-infinite channels in a porous two-dimensional body is solved by a method of direct statistical modeling.

Certain aspects of the parallelization of problems of the kinetic theory of gases solvable by direct statistical modeling on multiprocessor computers are discussed.

The best developed method for solving the complete Boltzmann equation without the a priori assumptions concerning the form of the distribution functions inherent in moment methods is direct statistical modeling.

The problem of steady supersonic condensation of a monatomic gas on a plane evaporating surface is solved in the Knudsen layer by the direct statistical modeling method.

Various multivariate statistical modelling techniques [latent class analysis (LCA), longitudinal LCA] were utilised to define the key features of PFS and describe its longitudinal characteristics.

Although LPL genotype did not significantly predict LV growth with ΔSBP in statistical modelling (LPL, P=0.14; ΔSBP, P=0.06), regression analysis indicated that LPL S447X genotype effect on ΔSBP accounted for only 20% of the effect on LV growth.

In differential GPS, once the observations have been screened for strong biases such as cycle slips, multipath, etc., the statistical modelling of the problem may still be unsatisfactory for some receiver-satellite pairs at certain epochs.

Statistical modelling using generalised additive models (GAM) revealed variations in activity patterns for two species.

Optimization of batch pyrite bioleaching with Sulfolobus acidocaldarius was performed using statistical modelling and experimental design.

In this paper, we point out the drawbacks of conventional statistic target models used in radar target modeling. It is usually difficult for us to approximate a real target by a conventional target model that has an analytical probability density function (PDF) expression, because there are very few parameters which can be used to approximate the real target's PDF by a theoretic target model's one. We suggest a new method of statistic modeling, in which the first nth central moments of real target's radar cross...

In this paper, we point out the drawbacks of conventional statistic target models used in radar target modeling. It is usually difficult for us to approximate a real target by a conventional target model that has an analytical probability density function (PDF) expression, because there are very few parameters which can be used to approximate the real target's PDF by a theoretic target model's one. We suggest a new method of statistic modeling, in which the first nth central moments of real target's radar cross section (RCS) are used to characterize the target, and the Legendre orthogonal polynomials are used to reconstruct the PDF of the target's RCS. The relationship between the coefficients of Legendre polynomials and the central moments of RCS are deduced mathematically. Through a practical computation example, the error-of-fit is shown as a function of the orders of Legendre coefficients. By comparing the errors-of-fit produced by both the new model and the conventional target models, we conclude that the new nonparametric method of statistic modeling is superior, for it makes the statistic modeling of radar target easier and more exact.

The present paper is a report on the research works completed during last five years by myself and my Ph. D. students at Peking University in connection with the problems in statistical time series and spatial series modelling. This paper is in two parts. In the first part, we shall give new results of estimating the orders and parameters of the stationary and monstationary ARMA models (Including stationary ARMA, ARUMA and the general ARMA models). The error terms are supposed to satisfy the martingale difference...

The present paper is a report on the research works completed during last five years by myself and my Ph. D. students at Peking University in connection with the problems in statistical time series and spatial series modelling. This paper is in two parts. In the first part, we shall give new results of estimating the orders and parameters of the stationary and monstationary ARMA models (Including stationary ARMA, ARUMA and the general ARMA models). The error terms are supposed to satisfy the martingale difference conditions which are weaker and more natural than supposing them to be i. i. d.. In the second part of this paper, we shall give new results about statistical spatial series modelling. Characterization of twodimensional ARMA models are given. A specific two-dimensional AR model(i. e. the quadrant Marker AR model) is found to be the exact two-dimensional counterpart of the classical onedimensional AR model. For the quadrant Markov AR model, we also give procedures of estimating their orders and parameters.

Based on the quadratic regression test and investigation of dynamic development of maize in experimental plot, the best farming measures for one ton per mu maize output were designed by means of statistical model and systematic simulation, which included density of 5.64~6.24 kiloplant/mu, 28.20~33.85kg/mu of topdressing pwre N per mu, P2O6 12.39~14.22kg/mu, K2O 23.16~24.80kg/mu, the plant highness and development of leaf area were shown to give the reason for field management, the analysis of the factors concerning...

Based on the quadratic regression test and investigation of dynamic development of maize in experimental plot, the best farming measures for one ton per mu maize output were designed by means of statistical model and systematic simulation, which included density of 5.64~6.24 kiloplant/mu, 28.20~33.85kg/mu of topdressing pwre N per mu, P2O6 12.39~14.22kg/mu, K2O 23.16~24.80kg/mu, the plant highness and development of leaf area were shown to give the reason for field management, the analysis of the factors concerning the maize output has shown, to obtain higher yield, the number of ear and grain should be stressed on the basis of reasonable plant density as well as kilograin. weight.