对humanether-a-gò-gò related genes(HERG)钾离子通道(钾通道)抑制剂,计算了表征分子组成、电荷分布、拓扑、几何结构及物理化学性质等特征的1559个分子描述符,采用Fischer Score(F-Score)排序过滤和MonteCarlo模拟退火法相...对humanether-a-gò-gò related genes(HERG)钾离子通道(钾通道)抑制剂,计算了表征分子组成、电荷分布、拓扑、几何结构及物理化学性质等特征的1559个分子描述符,采用Fischer Score(F-Score)排序过滤和MonteCarlo模拟退火法相结合从中筛选与HERG钾通道抑制剂分类相关的分子描述符.采用支持向量机(SVM)方法,分别以IC50=1.0、10.0μmol·L-1为分类标准,建立了三个分类预测模型.对367个训练集分子,用五重交叉验证,得到正、负样本的平均预测精度分别为84.8%-96.6%、80.7%-97.7%,其总的平均预测精度为87.1%-97.2%,优于其它文献报道结果.对97个外部测试集分子,所建三个模型的总样本预测精度在67.0%-90.1%之间,接近或优于其它文献报道结果.展开更多
An alternative option pricing method is proposed based on a random walk market model. The minimal entropy martingale measure which adopts no arbitrage opportunity in the market, is deduced for this market model and is...An alternative option pricing method is proposed based on a random walk market model. The minimal entropy martingale measure which adopts no arbitrage opportunity in the market, is deduced for this market model and is used as the pricing measure to evaluate European call options by a Monte Carlo simulation method. The proposed method is a purely data driven valuation method without any distributional assumption about the price process of underlying asset. The performance of the proposed method is compared with the canonical valuation method and the historical volatility-based Black-Scholes method in an artificial Black-Scholes world. The simulation results show that the proposed method has merits, and is valuable to financial engineering.展开更多
The theory of Monte-Carlo simulation method is that the event happened probability was estimated by happened frequency in the experiments. So it is very im-portant to ascertain the times of simulation. The probability...The theory of Monte-Carlo simulation method is that the event happened probability was estimated by happened frequency in the experiments. So it is very im-portant to ascertain the times of simulation. The probability and statistic theory are used to determine the optimal simulation times through strict mathematic reasoning. The study of two cases is made to show the availability of the presented method, and some relative problems were discussed.展开更多
Two methods currently available for evaluating the probability of Multiple Site Damage(MSD)occurrence were studied in this paper.One of the methods is a probabilistic analysis approach based on the statistical theory ...Two methods currently available for evaluating the probability of Multiple Site Damage(MSD)occurrence were studied in this paper.One of the methods is a probabilistic analysis approach based on the statistical theory and fatigue characteristics of each structural detail,and the other is an approach which defines the initial damage scenario by means of Monte-Carlo simulation,and multiple initial crack scenarios are randomly generated.A modified method based on the Monte-Carlo simulation was proposed in this paper,in which the random fluctuation of the stress was considered to give more accurate evaluation results.In the presented method,the probability of MSD occurrence in a structural element containing multiple details was calculated based on the Monte-Carlo simulation and the p-S-N curve of a single structural detail.Fatigue tests were accomplished using specimens containing 21-similar-details to obtain the fatigue life corresponding to MSD occurrence.Tests on single-detail specimens and static calibration tests were also conducted to get the basic fatigue properties of the material and the degree of stress fluctuation.The aforementioned three methods were compared and validated via the test results.The influence of the stress random fluctuation degree on the probability of MSD occurrence and influence of the distribution types on evaluating the MSD occurrence probability were discussed.展开更多
文摘对humanether-a-gò-gò related genes(HERG)钾离子通道(钾通道)抑制剂,计算了表征分子组成、电荷分布、拓扑、几何结构及物理化学性质等特征的1559个分子描述符,采用Fischer Score(F-Score)排序过滤和MonteCarlo模拟退火法相结合从中筛选与HERG钾通道抑制剂分类相关的分子描述符.采用支持向量机(SVM)方法,分别以IC50=1.0、10.0μmol·L-1为分类标准,建立了三个分类预测模型.对367个训练集分子,用五重交叉验证,得到正、负样本的平均预测精度分别为84.8%-96.6%、80.7%-97.7%,其总的平均预测精度为87.1%-97.2%,优于其它文献报道结果.对97个外部测试集分子,所建三个模型的总样本预测精度在67.0%-90.1%之间,接近或优于其它文献报道结果.
基金Funded by the Natural Science Foundation of China under Grant No.10571065.
文摘An alternative option pricing method is proposed based on a random walk market model. The minimal entropy martingale measure which adopts no arbitrage opportunity in the market, is deduced for this market model and is used as the pricing measure to evaluate European call options by a Monte Carlo simulation method. The proposed method is a purely data driven valuation method without any distributional assumption about the price process of underlying asset. The performance of the proposed method is compared with the canonical valuation method and the historical volatility-based Black-Scholes method in an artificial Black-Scholes world. The simulation results show that the proposed method has merits, and is valuable to financial engineering.
基金Supported by the National Natural Science Foundation of China(50274043)and Education Department of Hunan Province(04C544)
文摘The theory of Monte-Carlo simulation method is that the event happened probability was estimated by happened frequency in the experiments. So it is very im-portant to ascertain the times of simulation. The probability and statistic theory are used to determine the optimal simulation times through strict mathematic reasoning. The study of two cases is made to show the availability of the presented method, and some relative problems were discussed.
文摘Two methods currently available for evaluating the probability of Multiple Site Damage(MSD)occurrence were studied in this paper.One of the methods is a probabilistic analysis approach based on the statistical theory and fatigue characteristics of each structural detail,and the other is an approach which defines the initial damage scenario by means of Monte-Carlo simulation,and multiple initial crack scenarios are randomly generated.A modified method based on the Monte-Carlo simulation was proposed in this paper,in which the random fluctuation of the stress was considered to give more accurate evaluation results.In the presented method,the probability of MSD occurrence in a structural element containing multiple details was calculated based on the Monte-Carlo simulation and the p-S-N curve of a single structural detail.Fatigue tests were accomplished using specimens containing 21-similar-details to obtain the fatigue life corresponding to MSD occurrence.Tests on single-detail specimens and static calibration tests were also conducted to get the basic fatigue properties of the material and the degree of stress fluctuation.The aforementioned three methods were compared and validated via the test results.The influence of the stress random fluctuation degree on the probability of MSD occurrence and influence of the distribution types on evaluating the MSD occurrence probability were discussed.
