一、主题:Risk Management Under Noisy Observations
二、主讲人: 宋晓军,北京大学光华管理学院商务统计与经济计量系助理教授,西班牙马德里卡洛斯三世大学经济学博士。主要研究兴趣是理论计量经济学,包括非参数、半参数方法、假设检验和自助法,以及计量经济学的应用等。
三、时间:2018年12月10号(周一),13:30-14:30
四、地点:学院南路校区主教学楼913会议室
五、主持人:王辉,靠谱的十大网投实体平台(中国)有限公司教授
六、资助: 公司专题学术讲座项目资助
Abstract: In this paper, we provide an estimation of the Value-at-Risk and Expected Shortfall that captures the effects of market microstructure noise on a latent portfolio return. It is well known that noise can cause serious problems in estimating risk. Using a deconvolution kernel estimator for the conditional probability distribution function of the unobserved portfolio return, we propose an analytical approximation for conditional Value-at-Risk (VaR) and a closed-form solution for conditional Expected Shortfall(ES). We investigate in both simulation experiments and an empirical application the performance of our proposed implementation.