Research On The Correlation Of Flower Auction Based On Copula Function

The mainstream of Flower Industry in China is always traditional flower trade market. On November 12,2002, Kunming International Flower Auction Trading Center, the first flower auction market in our country, its formal operations indicates a new trading method come into being. In regard to this new trading method, We can research deal quantity, deal price and other transaction data. The flower deal quantity and deal price has some relationships, the deal quantity of same assortment and different rank has some relationships, the deal quantity of same rank and different assortment also has some relationships. Research these correlation can supply some advices to buyers, sellers and the market, thus buyers can get the most satisfied flower, sellers can get the largest profit, the auction market runs larger and better.The most traditional correlation analysis method is Granger causality analysis and linear correlation coefficient analysis, and the two kind of method is only suitable for linear correlation between variables, most variables have shown nonlinear relationship, the linear correlation coefficient to describe the correlation between them there will be some misleading. The copula function is a plurality of marginal distributions are connected together to form the joint distribution of the link function, by the correlation measure it derived in nonlinear monotone increasing invariant, can overcome the limitations of the linear correlation coefficient in use, there are more in the description of the correlation between financial variables and related structural advantages. Therefore, in this paper, using the copula function, we choose two variables flower deal quantity and flower deal price, to research the correlation between them, we also choose the same rank of carola and stars to research, remind the sellers pay attention to the trends of some variables, give the specific probability, apply quantizational level for the sellers, wish to raise their profit.Based on the introduction of Copula function theory and time series, to describe the correlation between flowers and flower auction sales price by establishing copula model, This paper uses the financial time series model to determine the marginal distribution, appropriate treatment through the analysis of sample data and using GARCH model to estimate the flowers and flower auction sales price earnings ratio of the marginal distribution, and to evaluate the effect of the fitting model; choose two normal Copula, two yuan t-Copula and Frank Copula in this paper to describe the relevant structural flower sales and flower auction price, through introducing the experience of Copula function, obtains the best two yuan t-Copula fitting effect conclusion.In order to measure the related structure and correlation of flower deal quantity and deal price as accurate as possible, we ultimately determine to use t-Copula-GARCH model in this paper, obtain that flower deal quantity and deal price have a negative correlation, and the same rank of carola and stars have a close positive correlation. The conclusions have theoretical and practical significance, provide some good suggestions for the traders in the flower auction market.