【摘要】 基于2018—2022年碳价、股价、外汇汇率和大宗商品小时单位数据,利用变系数模型和线性ARCH模型,构建了变系数风险度量模型,以资产相对成交量作为样条基函数,使用B样条分位数方法估计模型系数,测度高频金融数据的风险值,最后利用蒙特卡洛模拟验证了该估计方法的有效性和准确性。实证分析表明:资产相对成交量影响金融资产风险变动,B样条分位数方法的拟合成功率和误偏、误差值最小,拟合精度最高。更多还原

【Abstract】 Based on carbon price,stock price,foreign exchange rate,and commodity hourly unit data from 2018 to 2022,a variable coefficient risk measurement model is constructed by a variable coefficient model and a linear ARCH model.With the relative asset turnover as a spline basis function,the Bspline quantile method is utilized to estimate the model coefficients and measure the risk value of high-frequency financial data.Finally,Monte Carlo simulation is applied to verify the effectiveness and accuracy of the estimation method.Empirical analysis shows that the relative turnover of assets affects the risk changes of financial assets,and when the B-spline quantile method has the smallest fitting success rate,error bias,and error value,the fitting accuracy is the highest.更多还原