基于代价敏感学习的房地产企业财务预警模型分析-analysis of financial early warning model of real estate enterprises based on cost-sensitive learning.docxVIP

优秀毕业论文 精品参考文献资料 Abstract Financial crisis warning model’s establishment is on an ideal hypothesis that training set is absolutely right, which means that there is financial crisis in financial crisis enterprises and there is no potential financial crisis in financial health enterprises. But when we establish financial crisis warning model, how could we guarantee the correctness of training set’s category if we use BP neural networks method during training and simulation. It is fatal significance for financial crisis warning model based on the training set sample if there is wrong about the training set category, because this category mistake affects the accuracy of financial crisis warning model, which is crucial for the financial crisis warning model’s success or failure. Based on former scholar ’s research on financial crisis warning index system, the paper retains indexes extracted from the balance sheet, income statement and statement of cash flows, such as, solvency, profitability, operating capacity, ability to grow, ability to expand and index about cash flows and combines index about cooperate governance and audit opinion. The paper abandons these fuzzy indexes like strategic management, staff relationship and life cycle and index which can be calculated from existing index like EVA. Meanwhile, the paper puts forward land reserve index to highlight financial management characteristics of real estate companies and policy restrictions index to highlight China’s real estate market policy based on policy on finance, tax and deal registration. When the paper designs financial crisis warning process, BP neural networks category method is used to reflect the idea of cost -sensitive learning. After that, the paper establishes financial crisis warning model based on cost -sensitive learning and BP neural networks. The specific progress of the model is as follows: firstly, the paper standardizes 41 indexes of 9 aspects. Secondly, the samples are divided into trainin