【求指点】国内top4金融硕 准备研一开学后申请国外一流MF/MFE

seekingit

cheesechan 发表于 2016-4-18 19:02
容易留美的, 都是solid mathematical MFE,
數底不夠的, 不易申上.

非常感谢回复（之前在别的帖子里面看过层主的很多观点，受到很大启发）。
hard skills不算特别突出是因为其实本科阶段没有好好用功（只能自己再想办法弥补），但是hard>soft的原因是我soft skills是个人能力的短板……

advanced math/stat: time series analysis, multivariate statistics, econometrics(基本上算简化的回归分析吧），stochastic process.
ODE学了，PDE在考虑硕士期补课；Caluculus 有12个学分。
其他还有诸如operation research之类较基础的课。往届有申请的学姐，被小秘问学了这么多数学课为什么没有math minor，所以一定程度上衡量算是够用。
我比较了自己和NYU、Baruch项目的学生简历差距，觉得以现在的水平，应该是bottom quatile；但是也有本届数理情况和我差不多的同学申到tier 1.5的金工，所以我在认真考虑自学补数理基础；毕竟简历可能拿到面试（目前看了七八个tier 1 - tier 2的项目，基础课程完全无法不达到要求的只有Berkely），然后technical 面还是有机会的。
求指点，非常感谢！

cheesechan

seekingit 发表于 2016-4-20 22:25
非常感谢回复（之前在别的帖子里面看过层主的很多观点，受到很大启发）。
hard skills不算特别突出是因为 ...

- this level of math/stat should be similar to a minor level, and should be enough for MFE admission (but not a plus only), unless the MFE is a super mathematical one.
- get a C++ course will help quite much also if you don't have that before.
- numerical will be equally useful as PDE if not more. at the end you wont be required to solve stuffs analytically in practise.

seekingit

cheesechan 发表于 2016-4-20 23:22
- this level of math/stat should be similar to a minor level, and should be enough for MFE admissi ...

Many thanks~ 越深入了解越觉得MFE值得去争取一下。
1.之前有自学Python,现在会认真找C++的课程。
2.请问Numerical是指数学分析？我其实学过相关的course（接在单、多变量微积分后，为非数学系学生准备的），学得还不错。感觉PDE会跟难一点。
3.”won't be required to solve stuffs analytically in practice.“不太理解，是指更多的在做执行？

cheesechan

seekingit 发表于 2016-4-20 23:34
Many thanks~ 越深入了解越觉得MFE值得去争取一下。
1.之前有自学Python,现在会认真找C++的课程。
2.请问 ...

1. consider the Baruch quantnet one.
2. Numerical Analysis, and then Numerical Method for differential equations.
3. PDE teach the way of solving analytically and the theory behind.  Numerical method one will teach how to solve them numerically when analytical solution is not available.
In the practise of financial mathematics, analytical tractability is a nice properties for pricing models. But  product can be exotic, and thus the most common way is only rely on analytical tractability on calibration but not pricing, if possible. Not to mention that you can imagine that all those equations that can be solved analytically had be solved by researcher already, and thus understanding how things are solved numerically is also very important. In practise.......it is very likely you are either using it, repairing / maintaining it, or developing it......(it = tools), and this is the reason why people always say programming is important in this field.

seekingit

cheesechan 发表于 2016-4-20 23:40
1. consider the Baruch quantnet one.
2. Numerical Analysis, and then Numerical Method for differen ...

Get it.
Your sharing helps me a lot.

dereksong

那就更应该来MIT了哈哈，申的时候你可以说要做不那么quant的工作，来了直接选financial engineering track。同学啊，我们可是MIT，你只要想学，什么资源没有啊。。。master of finance是可以全校选课的，你天天跟phd上cs和数学都可以啊。我约了MIT Econ department副主任说想聊聊，直接就约上了啊，这可是MIT Econ啊同学。Sloan各种牛逼教授只要你有实力你跟他们做research都行啊。说MIT找工作不好，那是因为很多同学想找IBD，不是美本好学校的话去哪读master找美国IBD都很难啊。

上天赐我780

多选一些数理课 quit研究生干嘛 一边读 一边申请

Allen_Ssh

发表于 2016-04-13 19:11:46
本帖最后由 seekingit 于 2016-7-18 20:02 编辑

祝自己和大家好运~
...

看一下