The paradigm of economics is: 1. Literature review; Second, build your own mathematical model; Third, find data to test your model. The first part doesn't need much mathematical knowledge (but it needs a high level of English), the second part focuses on mathematical methods, and the third part focuses on the mathematical methods of measurement: 1. Learn the knowledge of preparation: set, relationship (equivalence, transmission, etc. ), total order, order, convex and concave, quasi-convex (concave). Know something about metric spaces. Understand the knowledge of quasi-concave function, concave function and differential calculus, as well as some knowledge of linear algebra. This knowledge will help you understand the content of advanced microeconomics, especially the proof of utility existence theorem, the understanding of general equilibrium and so on. It is particularly necessary to study the optimal behavior of economic individuals. Second, if you study macroeconomics, variational method and optimal control must be learned, otherwise advanced macroeconomics will be difficult to move. This requires knowledge of differential equations and a good economic foundation. Of course, if the method of differential equation is completely forgotten, it can also be realized by matlab software. However, economics needs more complicated relationships between variables. Software is software after all, and people don't understand it. Thirdly, in other related economic theories, random phenomena are often discussed, which requires some knowledge of mathematical statistics and probability theory, but personally, these theories are mostly concentrated in finance, which is rare in theoretical economics. 4. If you want to learn Yang Xiaokai's neoclassical economics, some knowledge of topology is essential, and you need to understand combinatorial mathematics. You also need to know the bang control of cybernetics and so on. Measurement method: 1. You must understand the reunification, otherwise you really can't say anything about economic research. To understand regression, we must understand linear algebra, probability theory and mathematical statistics (mainly hypothesis testing), otherwise we can't understand weird things such as F test and T test. What heteroscedasticity, sequence correlation and other issues in regression are not much to say. Eviews or spss software can be used mainly. Recommend spss because it is more intuitive. Second, the popular cointegration analysis (which will soon be out of date) seems to have to be learned. This requires a higher knowledge of linear algebra and mathematical statistics. Otherwise it is not easy to understand. Third, panel data analysis is the most popular now. The software used is stata8.0 and eviews5.0, otherwise programming analysis is needed. So if the knowledge of mathematical statistics and linear algebra is not good, these things cannot be said. Fourth, in addition to these, we need to know some statistical knowledge such as principal component analysis, factor analysis, cluster analysis and discriminant analysis. Of course, these analyses can be realized by spss. 5. Many scholars are paying attention to nonparametric statistics and nonparametric measurement. These methods are very helpful to test qualitative conclusions. You can look at it when you have time, and you need good knowledge of mathematical statistics. Economics has a high demand for mathematical knowledge, which is a headache. To tell the truth, knowing the above mathematical foundation may not be able to understand the relevant economics. Of course, I have limited ability and slow thinking. Ask the experts for more help. I weakly recommend several books: Mathematics: Fundamentals of Dynamic Optimization by Jiang Zhongyi, and his first book Fundamentals of Mathematical Economics suggests that students who don't know the basic knowledge of mathematics (that is, the number three in the postgraduate entrance examination) should also read it. His dynamic optimization is really good. Gong Liutang's Method of Dynamic Economics recommends reading the optimal control part, which may be the most comprehensive book in China (economics). The Method and Model of Economic Mathematics by Angel of Shanghai University of Finance and Economics is very close to the current economic research, and some basic concepts are also clearly stated. Measurement part: It is not recommended to read Econometrics by Mr. Li Zinai's Higher Education Press. Beginners are advised to read advanced econometrics. I recommend reading Sun's Econometrics, because there are many cases in it, so you can follow it. Factor analysis in statistics and the like can refer to a book introduced by spss, which does not require in-depth understanding. I am also depressed about panel analysis.