教师课堂教学质量的对应分析

教师课堂教学质量的优劣,直接影响着学生的学习效果和人才培养的质量,关系到学校的办学水平.科学分析研究教师的课堂教学质量,优化教学过程,实现教师和学生的全面发展,对提高整个高等教育的质量意义重大.以重庆第二师范学院经济与工商管理系为例,运用对应分析方法,分析了不同教师课堂教学的现状和特点,找出了影响课堂教学质量的因素,并针对各个不同类型的教师组提出分类引导,及时改进教学工作,提高课堂教学质量的建议.     

我国商业银行在聚合信用风险模型下的质押贷款风险度量探讨

传统的聚合信用风险模型在度量信用风险过程中,假定违约损失是给定不变的,但是近年的实际研究表明在实际金融市场中,违约损失是变化的。针对传统模型的这个不合理性。本文充分考虑了违约时损失程度的变化,引入信用等级转移矩阵和违约风险调整的短期利率来刘画这种变化,并利用总索赔量分布的Panjer递推算法给出了信用风险的度量．改进了文献[4,5]中概率生成函数算法的不足,对模型作了发展。     

Parametric multiple regression risk models: Connections with tariffication,especially in motor insurance

It is shown how the parametric multiple regression risk models of Albrecht (1983a) can be used for solving the problem of tariffication of e.g. an motor insurance portfolio.     

40Gb/s光时分复用传输光纤光栅补偿色散研究

用精密扫描掩模法写入宽阻带啁啾光纤光栅,掩模板背面两端各10％长度处镀有按4阶高斯函数透过率的膜,写入的啁啾光栅的时延纹波最大值为20ps。为减少写入光纤光栅的偏振模色散,研制了新的低偏振模色散光纤光栅补偿写入法。采用补偿写入法前的平均微分群时延为9．1406 ps;采用补偿写入法后的平均微分群时延为0．1521 ps。并利用低偏振模色散光纤光栅对40Gb／s光时分复用系统在普通G．652光纤传输122km的色散进行了补偿实验,功率代价为1．5dB。     

近红外光谱技术的小麦条锈病严重度分级识别

小麦条锈病是世界上影响小麦安全生产的一种重要病害。实现小麦条锈病不同严重度叶片快速、准确的分级识别,对于条锈病监测、预测预报和防治措施的制定具有重要意义。通过人工接种获得条锈病不同发病程度小麦叶片,选取8个不同严重度级别(1%,5%,10%,20%,40%,60%,80%和100%)叶片各30片和健康小麦叶片30片,利用近红外光谱技术分别获取光谱信息,共获得270条近红外光谱曲线,依据小麦叶片条锈病发病程度的不同,将其分为9个类别。从每个类别中随机选择7~8条光谱曲线作为测试集,共计67条,将剩余的203条光谱曲线作为训练集。利用定性偏最小二乘法建立小麦条锈病不同严重度叶片的定性识别模型。研究分析了不同光谱预处理方法、建模比(训练集:测试集)和建模谱区对所建模型识别效果的影响。结果表明,在4 000～9 000 cm-1谱区范围内,原始近红外光谱数据经中心化预处理后,建模比为3∶1时,采用内部交叉验证法建模,训练集和测试集的总体识别准确率分别为95.57%和97.01%,所建模型识别效果较好。表明基于近红外光谱技术进行小麦条锈病叶片严重度分级识别是可行的,为小麦条锈病的监测和评估提供了一种新方法。     

Pair-wise comparison experiment on subjective annoyance rating of noise samples with different frequency spectrums but same A-weighted level

A pair-wise comparison experiment on subjective annoyance rating of nine different noise samples is the focus of this paper. These noise samples collected from some associated facilities in urban residential areas are classified into three different types according to their frequency characteristics, and their A-weighted levels are equalized at 65 dBA. Based on the assessment of judging errors and statistical validity analysis of each testee’s evaluating matrix, invalid data are eliminated. The results show that the percentage of testees who chose noise samples containing more low frequency components had an obvious downside trend with the descending degree of annoyance from ‘the most annoyed’ to ‘the least annoyed’. The result indicates that noise which has low frequency characteristic is more annoying than noise of other frequencies when their A-weighted level are the same. Therefore, A-weighted level cannot scientifically assess noise annoyance which contains low frequency components mostly.     

Heat shrinkability of electron-beam-modified thermoplastic elastomeric films from blends of ethylene-vinylacetate copolymer and polyethylene

The heat shrinkability of electron-beam-irradiated thermoplastic elastomeric films from blends of ethylene–vinylacetate copolymer (EVA) and low-density polyethylene (LDPE) has been investigated in this paper. The effects of temperature, time and extent of stretching and shrinkage temperature and time have been reported. Based on the above data, the optimized conditions in terms of high heat shrinkage and low amnesia rating have been evaluated. Influence of radiation doses (0–500 kGy), multifunctional sensitizer levels (ditrimethylol propane tetraacrylate, DTMPTA), and blend proportions on heat shrinkability has been explained with the help of gel fraction and X-ray data. With the increase in radiation dose, gel fraction increases, which in turn gives rise to low values of heat shrinkage and amnesia rating. At a constant radiation dose and blend ratio, percent heat shrinkage is found to decrease with increase in DTMPTA level. Gel content increases with the increase in EVA content of the blend at a constant radiation dose and monomer level, giving rise to decrease in heat shrinkability. Heat shrinkage increases with the increase in percent crystallinity, although the amnesia rating follows the reverse trend.     

Phenomenological Modeling of Formic Acid Fractionation of Sugarcane Bagasse by Integration of Operation Parameters as an Extended Combined Severity Factor

In order to more conveniently simulate and optimize the solubilization of sugarcane bagasse components during formic acid (FA) fractionation, an extended combined severity factor (CSF_ext) was defined to integrate various operation parameters as a single factor. Two phenomenological models based on Arrhenius and Logistic equations were further used to describe the phenomenological kinetics. Different data-processing methods were compared to fit the severity parameters and model constants. Both Arrhenius-based and Logistic-based models show satisfying fitting results, though the values of Arrhenius-based CSF_ext (A-CSF_ext) and Logistic-based CSF_ext (L-CSF_ext) were somewhat different under the same fractionation condition. The solubilization of biomass components increased with CSF_ext, but two distinct stages could be observed with inflection points at A-CSF_ext of 42 or L-CSF_ext of 43, corresponding to bulk and residual solubilization stages, respectively. For the enzymatic hydrolysis of cellulosic solids, the highest initial enzymatic glucan conversion (EGC@6h) was obtained at A-CSF_ext of 39–40 or A-CSF_ext of 40–41; however, for a long hydrolysis period (72 h), relatively high glucan conversion (EGC@72h) was observed at A-CSF_ext of 42–43 or A-CSF_ext of 43–44. Post-treatment for deformylation with a small amount of lime could help to recover the cellulose digestibility.     

An approach to merit rating by means of autoregressive sequences

A posteriori ratemaking is widely applied in the premium calculation of property and casualty products, particularly in third-party automobile insurance, which usually uses a bonus–malus system for premium adjustment. The present paper suggests an alternative to common frameworks that are designed as random walks on graphs of mostly finite states representing premium levels. The proposed premium calculation model is governed by the policyholder’s claim history through a recursive equation. This new autoregressive scheme is structurally different from the ones in use. Relevant metrics that measure the system’s optimality are evaluated, partially in analytical form. Through a comparison with existing models and parameterisation from real-life data, the new model is put into context and its practical relevance is investigated.