Based On Matching Quantiles Estimation Model Factors Analysis Of The US Dollar Index

Since the breakup of the Bretton Woods system,countries have given up the fixed exchange rate system,and the volatility of the exchange rate in the international market has increased.With the deepening of economic globalization,the exchange rate risk has become one of the main risks.The dollar is the most important currency in the world monetary system.It has the dual attributes of the world reserve currency and the settlement currency,and its trend has a great influence on the world economy.The comprehensive value of the US dollar is measured by the US dollar index.It is one of the three important indexes that reflect the US and the global market economy vane.Considering that the US dollar index is so important,we can predict the trend of US dollar index through the analysis and analysis of the influencing factors of US dollar index.The article first introduces the concept of the dollar index,and sums up the factors affecting the dollar index into four aspects-the level of economic development in the United States,the macroeconomic policy of the United States,the rate of return brought 'by the dollar and the political factors.From these four points of view,the analysis of the volatility of the dollar index affects the fluctuation of the dollar index.Factors.Then the paper briefly introduces the traditional methods to study the influencing factors of the dollar index,namely the ordinary least squares regression model and the partial least squares regression model,and points out the shortcomings of the traditional research methods,that is,the non normal distribution characteristics of the partial least squares regression model to the dollar index and the distribution tail.The exception value of the Department can not be dealt with.On the basis of the first two chapters,the quantile matching estimation model which can solve the above problems is introduced.The theoretical basis of the quantile matching estimation model and the convergence property of the algorithm are introduced in detail,and the small sample properties of the quantile matching estimation model are proved by a set of simulated data.Finally,the paper makes an empirical study on the application of the factors affecting the dollar index to the partial least squares regression model and the quantile matching estimation model.From the level of economic development in the United States,we choose the United States GDP,CPI and the unemployment rate to compare the dollar index with the US dollar index.The difference between the benchmark interest rate,the long-term bond market and the stock market data between the EU and the United States,which is the largest US dollar rival,is compared with the dollar index.The empirical results show that the quantile matching estimation model is in the study of the factors affecting the dollar index,regardless of the fitting of the data in the sample.The result is better than the partial least squares regression model.