Non-Parametric Regression Methods

The nature of the financial time series is complex, continuous interchange of stochastic and deterministic regimes. Therefore, it is difficult to forecast with parametric techniques. Instead of parametric models, we propose three techniques and compare with each other. Neural networks and support vector regression (SVR) are two universally approximators. They are data-driven non parametric models. ARCH/GARCH models are also investigated. Our assumption is that the future value of Istanbul Stock Exchange 100 index daily return depends on the financial indicators although there is no known parametric model to explain this relationship. This relationship comes from the technical analysis. Comparison shows that the multi layer perceptron networks overperform the SVR and time series model (GARCH).     

基于改进的语音参数提取的线性预测

根据语音发生基本原理和线性预测编码原理，针对自相关法需要加窗从而降低了解的精度，而协方差法不能保证保证解所重构出系统的稳定性的特点，提出了一种改进的Cholesky分解的方法求解协方差方程组以产生声道模型参数的方法，实践证明，这种方法既保证了系统的稳定性，又提高了解的精度。     

几种常见光源特性的研究

本通过大学物理实验中常用光源特性的分析，说明了实验中光源使用的合理性。     

Theoretical analysis of the onset of liquid entrainment for slots of finite width

The onset of liquid entrainment during discharge from large reservoirs containing a stratified mixture of two immiscible fluids through a side slot of a finite width is considered theoretically. A previously reported analysis in which the slot was approximated as a two-dimensional line sink has been extended to account for the finite width of the slot. The model resulting from the present analysis is expressed in terms of two simple algebraic equations suitable for hand calculations. According to the present results, the ratio of the critical height to the slot width is dependent only on the Froude number. Numerical results show that the present model approaches the correct physical limits at low Froude numbers and it converges to the predictions of the previously reported simple model at high Froude numbers.     

石墨-六角氮化硼微晶混合物与水高压反应产物分析

 利用球磨法制备石墨-六角氮化硼微晶混合物，并在6.1 GPa、800~1 500 ℃条件下与水进行高压反应，以便研究用水作触媒合成B-C-N三元化合物的可能性。通过对反应产物的XRD、XPS谱分析发现：高压下随着温度的升高，反应产物中出现再结晶石墨，其晶化程度逐渐提高；但没有出现再结晶六角氮化硼，也未出现立方氮化硼。在球磨不充分条件下，石墨-六角氮化硼混合物的XRD谱没有完全弥散，它们与水高压反应时，能观察到石墨与立方氮化硼分别结晶的现象，但都没有形成B-C-N晶化结构。     

Interpolation on spherical geodesic grids: A comparative study

Most operational models in atmospheric physics, meteorology and climatology nowadays adopt spherical geodesic grids and require “ad hoc” developed interpolation procedures. The author does a comparison between chosen representatives of linear, distance-based and cubic interpolation schemes outlining their advantages and drawbacks in this specific application field. Numerical experiments on a standard test problem, while confirming a good performance of linear and distance-based schemes in a single interpolation step, also show their minor accuracy with respect to the cubic scheme in the more realistic simulation of advection of a meteorological field.     

����Լ���������������MinimaxԤ��

本文研究了独立随机变量之和的绝对矩的几个性质, 其中包括$\ep|X+Y|-\ep|X-Y|$的表达式, 这里$X$和$Y$是相互独立的随机变量.     

Pressure-Induced Shifts of R1 and R2 Lines of YAG:Cr3+

By means of improved ligand-field theory, the "pure electronic" pressure-induced shifts (PS's) and the PS's due to electron-phonon interaction (EPI) of R1 line and R2 line of YAG:Cr3 have been calculated, respectively.The calculated results are in very good agreement with the experimental data. It is demonstrated that the admixture of |t22(3T1)e4T2> and |t322E> bases in the wavefunction of R1 level of YAG:Cr3 and its change with pressure play a key role for the PS of R1 line. The behaviors of the "pure electronic" PS of R1 line and the PS of R1 line due to EPI are different. It is the combined effect of them that gives rise to the total PS of R1 line, which has satisfactorily explained the experimental results. The systematic analyses and comparisons between the feature of R1-line PS of YAG:Cr3 and the ones of three laser crystals (GSGG:Cr3 , GGG:Cr3 and ruby) have been made, and the origin of the difference between them has been revealed.     

应用加权列联表分析预测汛期（4─9月）洪涝趋势

本文根据汉寿县１９６０─１９８８年洪涝灾害资料，应用加权列联表分析法，与前期气候因子建立洪涝趋势预测模式。最后对１９８９─１９９０年洪涝试报，取得了满意的效果。