1. 招生政策
今年(2023Fall)的招生标准可能有所变化,Master 2nd year in Economic Theory and Econometrics (以下简称M2 ETE) 开始接受有四年本科背景的中国学生了,听学长姐说,以前的政策是如果没有硕士经历,要从M1开始读。
今年我们的cohort size==20,其中差不多1/3-1/2的学生是从M1升上来的,10个人是有一个TSE的Fellowship的教学金(10000欧元/年),对于这10个人来说,升为正式的PhD是只需要通过考核,而无须和其他人竞争。但另外10个人,如果想要读PhD,需要竞争5个名额,但通常来说,并不是所有人都想要读PhD的,所以基本上从M2 ETE升为正式的PhD并不会太难。
继续更课程安排(过于繁琐的描述,主要是为了给我自己存个档~)
每门课分成两部分,由不同的老师交。
2.1 Microeconomics:
Part 1: 基本上属于Intermediate Microeconmics (demand, supply, choice under uncertainty),还没有到MWG这本教材的难度。老师上课很丧,感觉他不喜欢教这一部分的知识,我们学起来也感觉挺没劲的(当然micro还是很有趣的)。有一份作业,难度适中,但是会比较繁琐?大家基本上都会自己找网上的lecture notes。这里推荐两个
a) https://sites.pitt.edu/~luca/ECON2100/2018Class/ (也包含了part 2的内容)
b) https://nmiller.web.illinois.edu//notes.html
Part 2: The main focus is general equilibrium! 会用到一些part 1学的demand & supply theory 但是不多。知识的深度有显著提高,似乎是phd level micro该有的难度了。过程大概是
1) Every allocation in the core is a paretp optimal allocation
2) Walrasian equilibrium is in the core thus pareto optimal (1st welfare theorem)
3) The existence condition of a Walrasian equlibrium
4) Every paretp optimal allocation can be acheived by a Walrasian equlibrium (2nd welfare theorem)
5) Incomplete market in dynamic setting.
老师讲的很好很清楚,特别喜欢上他的课。
2.2 Macroeconomics
Part 1: 讲了一个看起来很简单的模型,但是每一个necessary optimality condition (first order condition) 背后都是一个重要的结论,一个你必须记住的economics interpretation.
过程大概是
1) Production economy under central planner (endogenous income)
2) Endowment economy under central planner (exogenous income)
3) Endowment economy under two types of decentralization : time-0 trading and sequential trading
4) The equivalence between 2) and 3)
5) Production economy under decentralization-->The model used in business cycle anaylsis
6) Business cycle: key observations from the data (correlation and volatility of variables like output, hours worked, productivity etc.)
7) Calibration of the model in 5) with the data from 6)
8) Use the model in 7) to generate artificial data and compare with real world data from 6)-->business cycle puzzle
9) Some propositions on how to address the puzzle...(A huge literature on this...any one of them may become your midterm exam questions)
Part 2: 这一部分对我来说似乎更有趣,因为话题更多样化一些。Prof好强,给我们提供了很多intuition。然而我太菜,经常就是之能take for granted不能真正理解intuition……
1) Risk sharing (inherited in the models in part 1): more about testing...
2) Asset pricing key equation.
3) Equity premium puzzle (HJ bound, welfare costs of consumption fluctuation)
4) Business cycle accounting (wedges)
5) Misallocation (wedges)
6) Labor search (basic search model-->competitive search model).
2.3 Econometrics
一学期一共4次作业的econometrics真是让人又爱又恨。
Part 1: The basic tools in econometrics that every one should know. 老师是自己打的notes,但是怎么说,确实有点乱,我上课经常跟不上,只能求助于两本经典参考书,一本是Hansen,一本是Hayashi
大致如下
1) Linear least square
2) Asymptotic theory
3) 2 stage least square (IV)
4) M estimator (Maximum likelihood Estimation)
5) Genearlized method of moments (Method of moments)
6) Testing (Wald test, likelihood ratio test, Score test/LM test), size and power.
7) Test for endogeneity issue, test for identification issue
8) Weak instruments* (very briefly)
Part 2: Time series. 不要相信老师说的“这门课不会特别technical,重在给你带来一些intuition“。第一节课就把我吓懵了,这些词都是什么意思,我有必要知道么,如果有必要,是我能看懂的东西么……然后后面每节课都很难,我觉得是因为老师给了太多细节了,对于他来说这些细节可能并不复杂,但是对于初学者来说这些细节看起来就很吓人,以至于会让人迷失其中,不知道这节课的重点到底是什么。但是老师真的是一个非常非常非常好的人,我在复习周的时候几乎每天都去打扰他,经常是一个小时的office hour,甚至还有一个是在周六zoom,他还是特别耐心解答。当然参考书也不要相信lecture note上写的,认准Time series by Hamilton。
Syllabus 大致如下
1) The defintion of stationary, ergodic and mixing.
2) Wold representation theorem
3) The definition of auto covariance function, auto correlation function and partial correaltion function
4) Use the above 3 functions to characterize MA(q) AR(p) ARMA(p,q).
5) Auto covariance function and spectral density of a time series.
6) The canonical forms of the 3 types of representation and how to transform non-canonical form into canonical form.
7) Asymptotic theory
8) Testing/Inference with asymptotic theory
9) Estimation of AR (MLE), MA and ARMA (Kalman filter)
10) Non stationary time series such as random walk and intergrated of order 1. Need asymptotic distribution of the parameters of a non-stat time series for testing. (HARD)
2.4 Optimisation
选修课我选的optimisation
Part 1: 参考书是Optimization by Vector Space Methods。老师的时间规划不是很好,虽然他人很好玩,但是讲得很一般。看lecture notes和参考书才能懂得多一些。
Part 2: Calculus of variation-->Optimal control特别有意思的话题,主要是Daniel Liberzon的Calculus of variation and optimal control 这本书的前4章。通读下来神清气爽……(准确的说是蓬头垢面因为一动脑子就出油)
2.5 Game theory
旁听了大半学期之后没有继续了。但起码旁听的部分都挺有意思的。特别是Common knowledge 和public event 的部分。
4. 学术氛围
我们这一届cohort进入学的时候刚好就换了program director了,前director的宗旨是第一年no research and work hard on courses (you can work harder!!!)。 现任的director 的宗旨是work smart but not long (what am I supposed to if I am not smart QAQ)),read some papers。两种截然不同的风格吧……
每周都有很多seminar和talk,在 TSE event page上可以看到简介。每个seminar都会按照TSE的economics group分类,macro, econometrics and empirical economics, io, econ theory etc.
至于这个氛围和别的学校比怎么样,因为还没有去过其他学校,没有办法做counterfactual comparison嘻嘻:)
5. 杂七杂八 TSE building是普利策建筑奖的得主设计的,外墙的颜色用的是图卢茨经典的砖红。据说采光设计得很好,可惜我们ETE的办公室在不见天日的负一楼。
难得学校在市中心(有在内涵某些大学),离Place du Capitole 也不过10mins路程,去哪都挺方便的。