Research On Features Analysis Of Volatility Of Economic Cycle In China And Earning Warning Methods Under Uncertain Environment

With the development of computer, network and electronic technology in the 21 st century, the breadth and depth of information transmission are improved. Factors like nonlinearity, complexity and uncertainty have a great influence on the change of the economic cycle. For example, the US subprime mortgage crisis in 2008, causing lack of liquidity overworld. Since 2012, China’s GDP growth is slowing down, and macro economy has gone into a period of "new normal". So, in the new circumstances, are China’s economic cycle fluctuation characteristics of the present? How to make early warning in the new economic cycle under the condition of uncertainty?Economic cycle fluctuation is through a series of economic activities to transfer and diffuse, any single fluctuation process of an economic variable itself is not enough to represent the macroeconomic fluctuations in the overall process. Therefore, monitoring and characteristic analysis of the economic cycle fluctuations can’t only rely on GDP index, but alsosynthetically consider the production, consumption, investment, trade, finance, financial, employment in various fields such as climate change and mutual influence. The international and domestic researches on economic cycle fluctuation and early warning most focus on the recognition and judgement of economic cycle turning point. They commonly use GDP as the indicator at home and abroad to measure the economic cycle. They established the model according to GDP or adapt BP filtering, CF, filtering, etc. to eliminate trend, These methods are widely used in the study of the economic cycle, and has higher accuracy in the determination of economic cycle fluctuations. However, as more and more uncertainties in the economic development, we need to consider various social and economic factors in the economic cycle fluctuation early warning and be able to construct a model and method to deal with uncertainty.First of all, based on the literature of international and domestic economic cycle, we analysis the factors that affect the economic cycle at present stage, and try many kinds of existing advanced methods for quantitative monitoring of macroeconomic fluctuations in our country, and describe and analyze China’s economic cycle fluctuation characteristics in the new era is.Secondly, according to the characteristics of the economic cycle fluctuation, we use fuzzy soft set method to determine the economic cycle fluctuation. We contract and expanse the prosperity of the economic cycle semantic information using type ii fuzzy soft set, and at the same time, based on the definition of soft sets similarity, we use the proceeds to type 2 fuzzy soft set clustering, for China’s stage of economic cycle. In addition, we use trapezoidal fuzzy number to demonstrate the influence of the economic cycle variables to comprehensively calculate the degree value. Besides, according to Markov remige transmission probability, we construct fuzzy soft set degree value to measure the stage of economic cycle.Then we adopt soft smoothing to research on length and coefficient of the economic cycle. Length is an important tool to research on economic cycle. This paper defines the concept of soft smoothing, and process the tool of regression model soft smoothing to compare the lengths of economic cycle. We use soft smoothing to partly analysis soft set, aiming at further understanding characteristics of economic cycle and its efficiency of measurement. Considering the real situation of our country, we have found that volatility of economic cycle increases with the increase investment of fixed assets, while it decreases with the increase dependence of external trade. A large volatility in macroeconomic indicators may cause frequently economic volatilities,Finally, on the basis of characteristics analysis of economic cyle volatility, we use support vector machine(SVM) method and fuzzy regression method to make early warning annual and monthly climate index of the economic cycle. Support vector machine(SVM) method has the smaller error when dealing with nonlinear and high dimension problems.This paper shows a better result. Economic early-warning index reflects the short-term fluctuations in the economic cycle, our country and many countries in the world attach great importance to climate index. This article uses fuzzy regression analysis on economic early warning index to forecaste, and let the early warning index no longer limited to the light color.