国内环烷基原油减压渣油生产150BS光亮油工艺研究及收率预测

以国内几种典型环烷基原油减压渣油的轻脱油为原料,采用高压加氢中试装置进行150BS光亮油的生产工艺实验室研究,并对150BS光亮油的收率进行预测。研究结果表明,虽然环烷基原油来源不同,性质各有差异,采用相同的反应压力、体积空速、氢油比,得到黏度指数为80以上的150BS光亮油,其高压加氢处理反应温度均为382℃左右;采用全加氢工艺时,150BS光亮油的收率可以用-5%+100℃运动黏度数值×黏度指数数值/100的数学表达式进行预测。

基于偏最小二乘的Kriging代理模型在加氢裂化建模中的应用

提出了一种改进的代理模型方法 (Kriging with Partial Least Squares,KPLS)。该方法在Kriging模型的基础上引入偏最小二乘的思想,利用偏最小二乘方法构建新的Kriging模型的高斯核函数。将该模型应用于加氢裂化过程建模,有效地提高了航煤、柴油质量收率的预测精度。采用GLAMP(Global and local search strategy)优化算法对建立的KPLS模型进行优化,仿真结果显示航煤、柴油质量收率得到了显著提升。

Applications of nonferrous metal price volatility to prediction of China's stock market

The aim of the present work is to examine whether the price volatility of nonferrous metal futures can be used to predict the aggregate stock market returns in China. During a sample period from January of 2004 to December of 2011, empirical results show that the price volatility of basic nonferrous metals is a good predictor of value-weighted stock portfolio at various horizons in both in-sample and out-of-sample regressions. The predictive power of metal copper volatility is greater than that of aluminum. The results are robust to alternative measurements of variables and econometric approaches. After controlling several well-known macro pricing variables, the predictive power of copper volatility declines but remains statistically significant. Since the predictability exists only during our sample period, we conjecture that the stock market predictability by metal price volatility is partly driven by commodity financialization.

基于NSGA-II结合BP网络算法的乙醇制备C4烯烃优化模型

乙醇偶合制备C4烯烃工艺广泛应用于化工生产,探寻实现乙醇转化率与烯烃收率最大化的催化剂组合与温度条件具有重要意义。根据反应过程性能数据,分别运用多元回归与BP网络建立反应条件与烯烃收率的回归预测模型,结果表明BP网络模型对烯烃收率的预测值有较高的精度。建立多目标优化模型,基于BP网络良好的预测能力,采用NSGA-II算法进一步优化可得满足目标的催化剂与温度方案,此时乙醇转化率与烯烃收率分别为64.67%,35.921%。随后进一步分析比较NSGA-II算法和经典遗传算法分别优化BP网络模型获得的烯烃收率预测误差,验证了所用算法的精确性与有效性。

Risk-Neutral Dynamic of Forward-Looking Default Probabilities and Recovery Rates： Evidence From Credit Default Swap Prices of DOW30 Companies

In modem financial markets, the credit default swap （CDS） market has supplanted the bond market as the industry gauge for a borrower＇s credit quality. Therefore, it is very important to value CDS accurately by getting closer to more realistic pricing models. So far there have been no models for extracting forward-looking credit information to value CDS. In current practice, historical data is used in a credit default swap pricing model. One of the reasons was the difficulty when the market for credit derivatives was small, to extract forward-looking credit information such as recovery rates and default probabilities from traded securities. Since the CDS market has undergone rapid expansion in recent years, the possibilities of extracting forward-looking credit information have increased. Our work significantly extends Das and Hanouma （2009） where a flexible jump-to-default model was introduced to obtain implied recovery rates. We improve the flexible jump-to-default model where forecasted forward-looking hazard rates and recovery rates can be extracted using stock prices, stock volatilities and data from credit default markets to forecast CDS spreads. Instead of using exogenously assumed constant recovery rates and default probabilities from a credit rating agency, we use forward-looking hazard rates and recovery rates to price and forecast CDS spreads. We also compare out-of-sample market CDS spreads with our forecasted CDS spreads to check how well our model performs. Our model fit the market CDS spreads very well across all time to maturity CDS contracts except in some extreme cases when there is a big jump in CDS spreads.

Forecasting of Stock Returns by Using Manifold Wavelet Support Vector Machine

An admissible manifold wavelet kernel is proposed to construct manifold wavelet support vector machine(MWSVM) for stock returns forecasting.The manifold wavelet kernel is obtained by incorporating manifold theory into wavelet technique in support vector machine(SVM).Since manifold wavelet function can yield features that describe of the stock time series both at various locations and at varying time granularities,the MWSVM can approximate arbitrary nonlinear functions and forecast stock returns accurately.The applicability and validity of MWSVM for stock returns forecasting is confirmed through experiments on real-world stock data.