The Study Of Enterprise Period Gene Financial Early Warning Based On The Dynamic Time Warping Distance

Securities market is an important part of national economy, especially the stock market, is known as a barometer of macro economy, our country securities market experiences nearly 20 years of development, the scale is growing, which has a total market capitalization of $3.71 trillion, ranking the world’s fourth. The investment and financing function of the stock market can satisfy investors capital appreciation as well as enterprise capital accumulation to facilitate the needs of business for a long time. However, listed companies due to mismanagement, and inadequate liquidity into the financial crisis, will cause damage to the interest of securities investors, creditors, is not conducive to the stability of the securities market. According to the public disclosure of financial data of listed companies, analysis and forecasting the size of its financial risk in advance, will play a wake-up call to investors, optimize the investment decision-making, of the securities market investors, creditors and other stakeholders is important, to ensure long-term stability of the securities market.On the basis of combing the existing early warning model, this paper puts forward a kind of financial data of listed companies will be more cycle consisting of panel data discretization, timing, form each company specific time gene model, through dynamic distortion time to measure the similarity between the genetic distance, classifier was trained using data mining and classification to achieve the objective of the forecast of the listed company financial status. This model mainly includes three aspects: 1)make the company’s financial data of consecutive year as the time sequence, generate companies abstract period genes by data pretreatment;2) measure the similarity between the time gene, find out the best matching mode;3) through the classifier training data mining, analysis and forecast period genes corresponds to finances. Finally, using the data of A-share listed companies of five recently years for the empirical analysis, the results showed that genes during financial early warning method can better reflect the continuity of time financial crisis, and the similarity between the listed company, its forecast is more accurate.