Comparative Study Of The Methods Of Group-risk Crop Yield Ratemaking

As early as1949, Halcrow had regionaled crop yields on the concept of insurance, andthen, along with scholars in-depth study and practice of development in the United States andCanada, the program has been widely recognized. This kind of insurance plan for farmers, therelatively low rates of insurance products to meet the purchasing needs, and help it withstandsystemic risk; against insurers, the insurance product can eliminate moral hazard, reduceadverse selection, able to resolve the systemic risk to some extent, and the data readilyavailable, credible, can greatly reduce transaction costs; for the government, simply provideless premium subsidies, reduce the financial burden.The key crop yield insurance plans in the area of its ratemaking method often debateabout the rates determined by the parameters of law and non-parametric methods. Manyscholars are of the two methods were compared, the paper has also made attempts to try todispute the parametric method and non-parametric method to make a convincing answer.This article studies indicate parameter method may vary levelized years of losses andsevere underestimate risk in this regard due to the non-parametric method does not make anyassumptions about the distribution of production, can reflect more complete data reflect thetrue risk of accepting a non-parametric method applications. But there is also anon-parametric method may overestimate the risks, but also because of the amount of data,equations choose a stronger dependence and lack of stability. The parameter method is notsensitive to the amount of data but then, when the lack of data, parameter method may bemore stable results.Finally, since all the disadvantages of parametric method and non-parametric methods,based on consideration of coupling function is also regarded as a viable method first by somereasonable assumptions, adding a variety of meteorological factors affecting crop yield andthrough the raw data fitting crop production function, and then fully multi-output data throughcomputer simulations, and finally through the analog output of the data analysis, the results ofthe calculation of various rates. However, this method must be based on some of the morestringent assumptions, so the results lacked credibility. Currently the feasibility of choice may be concerned about the results of various methods, and appropriate weighting as the finalresult.