我国上市公司中并购目标公司的可预测性研究

发布时间：2018-12-18 10:35

【摘要】：随着我国资本市场的快速发展,上市公司的并购越来越活跃,并购目标公司的识别对于投资者、收购方和管理者来说都具有重要意义。在并购目标公司的可预测性研究方面,前人的研究往往年限较短,没有考虑行业差异产生的影响,且仅用一年或者某几年的预测准确性来说明并购目标公司的可预测性是片面的。本文对2001-2010年的样本逐年进行分析,每年都建立并购目标公司预测模型,并用下一年的并购事件对该预测模型进行检验,通过每年模型指标及准确率的变化,对我国控制权转移市场的并购目标公司可预测性问题做出相对全面、系统和客观的分析。本文采用通过实证分析得到以下结论： (1)目标公司具有盈利能力较差、偿债能力较差、营运能力较差、成长能力较差、获取现金能力较差及规模小的特点,对并购动机具有一定解释力；(2)通过单因素方差分析可以发现,在股权分置改革之前股权分散的上市公司容易被并购,而股权分置改革之后股权结构对成为并购目标公司的概率影响不显著；(3)通过三种方式建模并对预测模型进行检验,表明并购目标公司具有一定可预测性；(4)通过三种建模方式的对比,可以发现,采用逐年累计建模方式得到的预测模型拟合效果较好,也比较稳定,模型中较为稳定的指标包括总资产对数、经营产生的现金流量净额对数、扣除非经营性损益后的净利润对数及营运资金对数,这些指标的系数均为负。(5)通过三种建模方式的对比可以发现,按照各年样本建模的方式易受宏观因素影响,连续三年样本及逐年累计样本建模的方式建立的预测模型比较稳定,受宏观因素影响不明显,且逐年累计样本建模方式的预测结果及模型的稳定性最好,由此表明随着建模样本数量的增加可以有效降低宏观因素的影响。
[Abstract]:With the rapid development of China's capital market, M & A of listed companies is becoming more and more active. The identification of target companies is of great significance to investors, acquirers and managers. In the research of the predictability of M & A target companies, previous studies tend to have a short period of time and do not take into account the impact of industry differences. And the prediction accuracy of only one year or several years to explain the target company M & A predictability is one-sided. In this paper, the samples of 2001-2010 are analyzed year by year, and the forecast model of target company is established every year, and the forecast model is tested by the next year's M & A event, and the change of the index and accuracy of the model is made every year. This paper makes a relatively comprehensive, systematic and objective analysis on the predictability of M & A target companies in the market of control transfer in China. In this paper, the following conclusions are obtained by empirical analysis: (1) the target company has the characteristics of poor profitability, poor solvency, poor operating capacity, poor growth capacity, poor ability to obtain cash and small scale. It has certain explanatory power to the motive of M & A; (2) through the analysis of variance of single factor, it can be found that the listed companies with dispersed shares are easy to be merged before the split share structure reform, but the ownership structure has no significant effect on the probability of becoming the target company after the split share structure reform; (3) through modeling and testing the prediction model in three ways, it shows that the target company of M & A has certain predictability; (4) through the comparison of the three modeling methods, it can be found that the prediction model obtained by the cumulative modeling method has good fitting effect and stability, and the more stable indexes in the model include the logarithm of total assets. The logarithm of net cash flow generated by operation, the logarithm of net profit after deducting non-operating profit and loss and the logarithm of working capital, the coefficients of these indexes are all negative. (5) through the comparison of three modeling methods, we can find that, According to the method of sample modeling in each year, it is easy to be influenced by macroscopical factors. The prediction model established by the method of sample modeling for three consecutive years and accumulative sample year by year is relatively stable, but not obviously influenced by macro factors. Moreover, the prediction results and the stability of the model are the best, which shows that with the increase of the number of modeling samples, the influence of macroscopic factors can be effectively reduced.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F271;F832.51;F224

【参考文献】