Reputation of Max Planck Society vs. Ivy League PhD

[MaxML]

Hi all,
I am currently working on my PhD in theoretical physics at a Max Planck Institute in Germany with a focus on mathematical modelling using machine learning. Meanwhile I have got interested in the financial sector given that also there you can do a lot of maths and now I wonder if I have chances to get a job as a quant in investment banking or in a hedge fund with an education at a German technical university and a PhD from a Max Planck Institute.

I have the impression that the Max Planck Society isn't as present in people's minds as for example Ivy League or Oxbridge universities are although being on par from a scientific point of view:

The titans: Institutional rankings by output and citations
2017 tables: Institutions - Nature & Science | 2017 tables | Institutions - <i>Nature</i> & <i>Science</i> | Nature Index


What is your opinion?

 

[Daniel Duffy]

Quality, not quantity.

The number of published articles is an indicator of what?

 

[MaxML]

Sure, the number of total publications should not measure the research quality of the institute. The second table in the first reference represents the number of total citations. Considering the ratio of #citations and #papers the Max Planck Society looks comparable with the other top schools. However, one should maybe compare by subject as there are subjects generating higher rates of citation (see text of first reference) and also by other factors.

 

[Daniel Duffy]

Seems everyone is doing ML these day all over the whole world (hope it's not Tulip Mania).

What specific skills have you learned that you could apply in finance? Have you done (preliminary) investigations yet? I am not an interviewer but it would be one of my 1st questions for you.

 

[MaxML]

Which of my skills is of value is actually also a question to me at the moment. I haven't really started dealing with the finance stuff yet.

The goal of my PhD projects has been identifying specific relationships and trends from existing data to both make predictions and understand the relationships on the one side, on the other side approximate mappings which are computational demanding.
Besides neural networks and kernel based methods (kernel ridge regression, Gaussian process regression, support vector machine) I have worked a lot on compressed sensing.

One skill I would mention is the ability to tackle data science problems from scratch.
I have mainly used python (daily) and also some fortran, unfortunately .

 

[Daniel Duffy]

The book "Advances in Financial Machine Learning" by De Prado might be useful. Examples in Python.

In general, learning "bigger" languages such as C++11/14/17, C# or Java is an asset. Python is useful for scripting etc.