CAS39号文件实施后公允价值分层计量的价值相关性研究

发布时间：2018-05-08 12:07

本文选题：公允价值 + 分层计量　；参考：《西南交通大学》2017年硕士论文

【摘要】：2008年的金融危机使公允价值计量属性差点退出历史舞台,幸亏国际学者们对公允价值计量做出一系列的实证检验后显示公允价值计量可以提供比历史成本计量更加可靠的财务信息,力挺了公允价值计量的存在意义。之后随着经济的复苏,该计量属性渐渐被市场所认可,被企业普遍运用,特别是金融行业的上市公司。此外,随着我国市场经济的不断成熟,财务信息使用者对企业会计准则提出了新的要求,他们希望能够得到更加可靠的会计信息,以帮助其做出更加有效的投资决策,加上在国际会计准则的推动下,2014年我国财政部发布了 CAS39号文件,这是我国针对公允价值计量属性做出的最新的要求,其中未了响应国际号召,该文件增加了当前国际上普遍使用的公允价值层级理论。即财政部规定,2014年7月开始各行各业应对公允价值金额进行分层披露,以增强财务会计信息的可靠性,并以此来达到信息使用者对其提出的新要求。本文针对公允价值分层计量,在查阅大量相关资料后,通过改进的Ohlson模型,以2014-2016年上海交易所和深圳交易所的金融行业上市公司为研究对象,对CAS39号文件实施后的以各层级公允价值计量的资产和负债与股价的相关性进行了实证研究。其中,由于CAS39号文件是在2014年才开始实施,考虑到数据规模对研究结果的影响,将上述研究样本的半年报数据纳入研究范围。本文选取4月和8月最后交易日的股价作为因变量P,自变量为金融行业上市公司年报和半年报中公允价值资产和负债的层级信息的每股数,分别为:第一层级每股公允价值资产FVA1it、第二层级每股公允价值资产FVA2it、第三层级每股公允价值资产FVA3it、第一层级每股公允价值负债FVL1it、第二层级每股公允价值负债FVL2it、第三层级每股公允价值负债FVL3it,其余的参变量有公允价值总资产金额FVA1it,公允价值总负债金额FVL1it,账面价值总资产与总负债的差额NFVALit以及扣除非经常性损益后的基本每股收益NIit。在数据整理过程中,剔除了数据不完整的样本,得到220个有效样本。在实证部分使用的是Stata数据处理软件,首先对样本行业分布、样本规模以及全部变量做了描述性分析,然后对模型进行了多元回归分析,最后将因变量P换成4月和8月最后交易日前的15个交易日的均值进行稳健性检验。最终说明CAS39号准则的实施情况比较良好,各层级公允价值计量的资产负债与股价的相关性如下:第一,单独检验资产,第一层公允价值计量的资产具有价值相关性,而且是正相关,第二、三层次公允价值计量的资产价值相关性具有滞后性,并且负相关,各层级没有逐层降低趋势。第二,单独检验负债,各层次公允价值计量的负债均具有价值相关性并且负相关,逐层降低趋势不稳定。第三,资产负债整体检验,第一层级公允价值计量的资产和第一二层级负债均具有价值相关性,第二层级资产和第三层级负债价值相关性不稳定,第三层级的公允价值资产不具有相关性,整体检验中各层级也没有呈现逐层降低趋势。这可能是由于第三层级的数据披露不规范以及第三层次本身数据太小所致,也说明企业需要加强估值技术的完善。
[Abstract]:The financial crisis of 2008 made the measurement attribute of fair value out of the historical stage. Fortunately, the international scholars have made a series of empirical tests on the fair value measurement and show that the fair value measurement can provide more reliable financial information than the historical cost measurement, and strengthen the significance of the fair value measurement. After the recovery, the measurement attribute is gradually recognized by the market and is widely used by the enterprises, especially the listed companies in the financial industry. In addition, with the continuous maturity of the market economy in China, the users of financial information have put forward new requirements for the accounting standards of enterprises. They hope to get more reliable accounting information to help them make more effective. The investment decision, coupled with the promotion of international accounting standards, was issued by the Ministry of finance of China in 2014, which is the latest requirement for our country's fair value measurement attribute, which has not responded to the international call. This document adds to the current universal use of fair value level theory in the world. That is, the Ministry of Finance stipulates that 2014 In July, each industry began to disclose the fair value of the fair value in order to enhance the reliability of the financial accounting information and to achieve the new requirements of the information users. In this paper, the fair value stratified measurement, after consulting a large number of relevant data, through the improved Ohlson model, 2014-2016 years Shanghai exchange and deep. The financial industry listed company of Shenzhen stock exchange is the research object. The empirical study on the correlation between assets and liabilities measured at the level of fair value of each level after the implementation of document CAS39 was carried out. In this case, the CAS39 document was implemented only in 2014. Considering the impact of the data pattern on the results of the research, the above research samples are taken into account. The share price of the last trading day in April and August is selected as the dependent variable P, the independent variable is the level of the level information of the fair value assets and liabilities in the annual and semi annual reports of the financial industry listed companies and half a year, respectively: the first level of fair value per share FVA1it, second levels of fair value per share. FVA2it, third levels of fair value per share FVA3it, the first level of fair value per share FVL1it, second level share Fair Value debt FVL2it, third level per share Fair Value debt FVL3it, the rest of the variables have fair value total assets FVA1it, fair value total liability amount FVL1it, total book value assets and total The difference in debt NFVALit and the basic earnings per share after deducting non recurrent gains and losses NIit. in the process of data sorting out of the data incomplete samples, get 220 effective samples. In the empirical part, the use of Stata data processing software, first of the sample industry distribution, sample size and all variables to do descriptive analysis, Then the multiple regression analysis was carried out. Finally, the variable P was changed to the average of the 15 trading days before the April and August last trading days. Finally, the implementation of CAS39 standard was better. The correlation between the balance of assets and liabilities and the stock price of all levels of fair value measurement was as follows: first, test assets alone, first. The assets of the fair value measurement of the layer have value relevance and positive correlation. The second, third level fair value measurement of asset value has a lag, and negative correlation, and there is no layer by layer to reduce the trend. Second, the liabilities are separately tested, and the liabilities of all levels of fair value are both of value relevance and negative correlation, Third, the overall inspection of the assets and liabilities, the first level fair value of the assets and the first two level liabilities have value relevance, the second level assets and the third level debt value correlation is unstable, the third level of fair value assets are not related, the overall inspection of the various levels is also not. There is a layer by layer reduction trend. This may be due to the third levels of data disclosure is not standardized and the third level of the data is too small. It also indicates that the enterprise needs to strengthen the improvement of valuation technology.

【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F830.42;F832.51

【参考文献】