Best Courses to take for MS Statistics (BuySide Interest)

[jsmith__]

Hi All,

I am currently enrolled in a MS Statistics program and I have the option to take three electives. I'm interested in joining the buyside and working as a quant when I graduate. For my electives, I know Applied Time Series Analysis is a must so I will go with that. Not sure what to take for my other two classes. Some options I were thinking could be Applied Econometrics I & II, Applied Bayesian Analysis, Applied Multivariate and Longitudinal Data Analysis, Neural Networks and Deep Learning. Thanks!

 

[MikeLawrence]

Bayesian pops up a lot, some buy side firms claim to build their methods around it. That would be a good one. Time series is needed, yes; ideally you cover through ARCH/GARCH methods. Applied Multivariate might be the most useful of the ones remaining, but NN and DL are hot topics and could be interesting.
I assume you have at least one course in computational methods already, if not then that needs to be added to the list.

Edit: Listen to Mr. Abbott over me.

 

[Ken Abbott]

Hi All,

I am currently enrolled in a MS Statistics program and I have the option to take three electives. I'm interested in joining the buyside and working as a quant when I graduate. For my electives, I know Applied Time Series Analysis is a must so I will go with that. Not sure what to take for my other two classes. Some options I were thinking could be Applied Econometrics I & II, Applied Bayesian Analysis, Applied Multivariate and Longitudinal Data Analysis, Neural Networks and Deep Learning. Thanks!

Consider optimization: LP, QP, NLP.

 

[marcusaurelius]

Consider optimization: LP, QP, NLP.

Any recommendations of textbooks? I saw recommendations for other topics on QN, but not for optimization... Any tips appreciated!

 

[Daniel Duffy]

This is not bad

https://www.amazon.com/Numerical-Optimization-Operations-Financial-Engineering/dp/0387303030/

and this has LOTS of examples

https://www.amazon.com/Schaums-Outline-Operations-Research-Richard/dp/0070080208/

 

[Daniel Duffy]

This is not bad

https://www.amazon.com/Numerical-Optimization-Operations-Financial-Engineering/dp/0387303030/

and this has LOTS of examples

https://www.amazon.com/Schaums-Outline-Operations-Research-Richard/dp/0070080208/

And havinng understand it all, then Python libraries give implementations.

 

[marcusaurelius]

And havinng understand it all, then Python libraries give implementations.

Yep, makes sense! Thanks for the advice!

 

[Daniel Duffy]

Yep, makes sense! Thanks for the advice!

I sometimes go directly to a library as a way to learn a new topic. It breaks the ice as it were.
Goofing around throws up unexpected results, which can be good.

 

[marcusaurelius]

Good idea!

 

[Ken Abbott]

Any recommendations of textbooks? I saw recommendations for other topics on QN, but not for optimization... Any tips appreciated!

Probably better to use an online resource to get started and then try some examples in Matlab.

 

[Daniel Duffy]

A of Machine Learning, calibration etc. demands a good knowledge of optimisation.

Nothing takes place in the world whose meaning is not that of some maximum or minimum.

- Leonhard Euler

 

[Harold_M]

Just to hop on the thread, is Probabilistic Modeling a valuable skill in the QR sector? Topics like Generalized Linear Models, Exponential Family, Bayesian Nonparametric etc? Wondering if I shall take the elective in Probabilistic Modeling and Machine Learning, or just stick with the Time Series Analysis.

 

[jsmith__]

Consider optimization: LP, QP, NLP.

Thank you for this. Also, just curious on why you recommended optimization (LP, QP, NLP)?

 

[Ken Abbott]

Probably better to use an online resource to get started and then try some examples in Matlab.

BUT: I found it very helpful to run the optimizations in a spreadsheet or programming language before you run them in a math/stat package

 

[marcusaurelius]

BUT: I found it very helpful to run the optimizations in a spreadsheet or programming language before you run them in a math/stat package

Noted

 

[Daniel Duffy]

A good test: solve an optimisation by hand, i.e. without a computer, e.g. graphically or high-school algebra.
Low-dimensional problems.