摘要
本文基于状态空间模型构建财务危机的动态预警系统,应用卡尔曼滤波计算模型的时变参数。并以数据统计为基础,根据上市公司绩效的动态特点,提取公司状态由好向坏转变,危机由轻至重的两个阈值分割点。研究从CCER数据库选取264家上市公司,30个财务指标,时间跨度为1994-2008年、合计2724组年度观测数据作为研究样本进行时序的分析与检验。实证结果表明状态空间模型能够有效刻画企业财务状况随时间序列的累积变异,实现动态预警系统的递归更新和实时预测。
This paper studies on empirical analysis of China listed companies based on financial distress early warning state space dynamic mode. Kalman filtering is used to calculate the time - varying parameters. According to the dynamic process of companies'performance, the financial states of a company are divided into three phases, and are identified two thresholds according to financial statistics. We use CCER database and select 264 China listed companies as study samples. We chose 30 financial ratios from 1994 to 2008, including 2724 year - observation data to empirical study. It is tested to plot effectively the the time series cumulative variance of companies'financial situation, and to realize the dy- namic early warning and real - time forecasting.
出处
《中国软科学》
CSSCI
北大核心
2013年第4期140-147,共8页
China Soft Science
基金
国家自然科学基金重点项目(71031003)
教育部人文社科项目(11YJC630188)
上海市科研创新项目(13YZ083)
上海海事大学基金项目(20110044)
上海市重点学科建设项目(S30601)
关键词
财务危机动态预警
状态空间模型
卡尔曼滤波
三阶段阈值
financial crisis early dynamic warning
state space model
kalman filtering
three-stage threshold