Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Careers - Employment
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions



Reply
 
Thread Tools Search this Thread Display Modes
  #101  
Old 02-11-2018, 08:27 PM
ThereIsNoSpoon ThereIsNoSpoon is offline
Member
CAS SOA
 
Join Date: Sep 2014
Studying for PA,FAP
College: When the smog clears, _ _ _ _
Favorite beer: PABST! BLUE RIBBON!
Posts: 452
Default

Quote:
Originally Posted by MathStatFin View Post
These MS in analytics programs tend to be s*it. There are only a few good masters programs in the United States. You'll be studying with people who have weak backgrounds and it'll be dumbed down. If you're just looking to pick up a masters without too much challenge, then go ahead and enroll. I would think that you could do a lot better considering you were able to pass actuarial exams.
and which are the few good ones?

Frankly I'm considering a certain few European masters programs instead because they've got some robust research going on while still not blowing a hole clean in my pocket. I was surprised to know that Zurich's statistics program costs less than a thousand francs a semester!
__________________
Spoiler:

------------------------------------------
P FM MFE C MLC VEE Economics VEE Applied Statistics VEE Corporate Finance PA FAP

Want to connect on LinkedIn? PM me!
Reply With Quote
  #102  
Old 02-12-2018, 11:15 AM
Neutral Omen's Avatar
Neutral Omen Neutral Omen is offline
Member
SOA
 
Join Date: Oct 2011
Favorite beer: Kwak
Posts: 6,990
Default

Quote:
Originally Posted by ThereIsNoSpoon View Post
and which are the few good ones?
Clarinetist already said: CMU, Cornell, Stanford.
__________________
Spoiler:


Quote:
Originally Posted by mathmajor View Post
you are a visionary
Reply With Quote
  #103  
Old 02-13-2018, 07:51 PM
MathStatFin's Avatar
MathStatFin MathStatFin is offline
Member
Non-Actuary
 
Join Date: Jul 2014
Posts: 4,676
Default

Quote:
Originally Posted by ThereIsNoSpoon View Post
and which are the few good ones?

Frankly I'm considering a certain few European masters programs instead because they've got some robust research going on while still not blowing a hole clean in my pocket. I was surprised to know that Zurich's statistics program costs less than a thousand francs a semester!
Zurich is one of the best Universities for math/stats/cs. It's probably top 3-5 in Europe. Swiss schools have a reputation for being pretty tough/rigorous. I'm sure their program is excellent. It's just not clear how you would fare with a degree from Zurich when applying for jobs in the US. In terms of academics and costs, I'm pretty sure Zurich is better than anything the US is offering.
Reply With Quote
  #104  
Old 02-13-2018, 08:15 PM
MathStatFin's Avatar
MathStatFin MathStatFin is offline
Member
Non-Actuary
 
Join Date: Jul 2014
Posts: 4,676
Default

Quote:
Originally Posted by extrovertedactuary View Post
Really? Financial services is strongly represented, but it shows the names of the employers that made offers and it bolds the names of employers that actually hired graduates of the program.

http://analytics.ncsu.edu/?page_id=13160

The chart provided says 25% of graduates of the program work in financial services. 20% work in consulting. Even if I granted you the false assumption that 100% of those working in consulting work in financial services, still fewer than half the graduates would be working in financial services.

http://analytics.ncsu.edu/?page_id=248

Did you even read the website other than the curriculum?
It shows pictures of recent graduates of the program along with their past employers, past education, new employer after graduating, new location, and new position title. And it's not just one or two. It shows this information for more than 20 recent graduates.



You're using way too much anecdotal evidence by posting one poster's Reddit comment. Although I agree with the point that more traditional degrees are more widely accepted as they've been around longer, I think you're really pushing a false narrative here.

I actually didn't disagree with your comment that there are a lot of shitty data science/analytics programs, because I agree with that. I just strongly disagree with your comment that there are only a few good masters programs in the US. Unless by a few you mean way more than I'm reading into, I think your comment is just way off.

In regard to the poster who is applying to the masters program in analytics, I think he has very little to lose in that he's applying to a very financially affordable program at a respected school and already is working as an actuary. I don't see how it can hurt him to add some analytics skills to his toolbox.

Also, I've browsed some of the recruiter websites, and a bunch of roles have mentioned that they would like an actuary who has a masters or higher in a quantitative field. Having the analytics masters could open up some of these opportunities for the poster.

I personally know a few people outside of my current company who have masters degrees in business analytics, and one works for the DoD, another for GE, and another for Apple.

If the main point of your original post was to highlight that a lot of these programs are scams, then I agree with that point 100%. However, far more than a few masters programs in the U.S. are good if by good we mean cost, career placement rates, companies that hired, and starting pay.
I was merely trying to provide a data point related to what I said. Think about this for a sec. The majority of data scientists have traditional grad degrees and learned all this "data science" stuff on their own. It's like how a non AS grad who passed exams would look down on an actuarial science grad. With time you will have more and more people with MS data science/analytics getting jobs and they will probably tend to hire people with the same credentials (just like AS grads might prefer to hire other AS grads). It's basically just history repeating itself. I expect current data scientists, the ones doing more complicated stuff, will rebrand themselves as "ML scientists" or "ML researchers" (something more granular).

Two other points I want to make:

1. I checked the student profiles for the NCSU program and I would say that at least half the students come from a non-stem background. There's even one student with a degree in theatre. A lot of people with business degrees. That's a little bit of a red flag imo.

2. Starting salaries aren't that important. It's better to focus on the long term especially when you're already an experienced actuary. these programs are teaching the hottest/latest frameworks/languages/tools which may not be as relevant in a couple of years.
Reply With Quote
  #105  
Old 02-14-2018, 10:11 AM
Neutral Omen's Avatar
Neutral Omen Neutral Omen is offline
Member
SOA
 
Join Date: Oct 2011
Favorite beer: Kwak
Posts: 6,990
Default

Quote:
Originally Posted by MathStatFin View Post
Zurich is one of the best Universities for math/stats/cs. It's probably top 3-5 in Europe. Swiss schools have a reputation for being pretty tough/rigorous. I'm sure their program is excellent. It's just not clear how you would fare with a degree from Zurich when applying for jobs in the US. In terms of academics and costs, I'm pretty sure Zurich is better than anything the US is offering.
Is the Zurich curriculum in English? Because German will take a few years to learn.
__________________
Spoiler:


Quote:
Originally Posted by mathmajor View Post
you are a visionary
Reply With Quote
  #106  
Old 02-14-2018, 01:01 PM
whoanonstop's Avatar
whoanonstop whoanonstop is offline
Member
Non-Actuary
 
Join Date: Aug 2013
Location: Los Angeles, CA
Studying for Spark / Scala
College: College of William and Mary
Favorite beer: Orange Juice
Posts: 5,830
Blog Entries: 1
Default

Quote:
Originally Posted by MathStatFin View Post
The majority of data scientists have traditional grad degrees and learned all this "data science" stuff on their own.
Yeah, to be honest it doesn't matter what degree you have if you're able to pick stuff up on your own, but when you condition on a traditional degree the chance of that goes up. In the end, I just don't have faith that spoon-fed programs will produce the highest tier people for the long term.

Quote:
Originally Posted by MathStatFin View Post
1. I checked the student profiles for the NCSU program and I would say that at least half the students come from a non-stem background. There's even one student with a degree in theatre. A lot of people with business degrees. That's a little bit of a red flag imo.
Noticed the same thing myself. Huge red flag as I have little faith the majority of those people could handle any technical rigor. Even better, note that they point out under education their "SAS credentials", which I'm pretty sure every candidate has. Must be a pre-requisite for the program?

http://analytics.ncsu.edu/reports/MS...t_Profiles.pdf

Quote:
Originally Posted by MathStatFin View Post
2. Starting salaries aren't that important. It's better to focus on the long term especially when you're already an experienced actuary. these programs are teaching the hottest/latest frameworks/languages/tools which may not be as relevant in a couple of years.
The salaries are better than EL actuarial salaries but not as high as EL data science.

-Riley
__________________
Reply With Quote
  #107  
Old 02-14-2018, 01:34 PM
nonlnear nonlnear is offline
Member
Non-Actuary
 
Join Date: May 2010
Favorite beer: Civil Society Fresh IPA
Posts: 29,619
Default

Quote:
Originally Posted by whoanonstop View Post
Yeah, to be honest it doesn't matter what degree you have if you're able to pick stuff up on your own, but when you condition on a traditional degree the chance of that goes up. In the end, I just don't have faith that spoon-fed programs will produce the highest tier people for the long term.
Especially given that the current state of the art in machine learning algorithms is... the ensemble. Which amounts to "Throw lots of different pieces of crap against the wall to maximize expected stickiness." In a world where that is the current state of affairs, a quickie certificate/masters where students get taught the latest batch of stickiest models might not be producing the right kind of understanding for the advancement of the field as a whole. What you need is people who understand things like why the Kadison–Singer problem solution matters, not people who understand that turbo boosted green trees are 5% better than anisotropic Haarlets at determining if a pink fluffy unicorn is in fact dancing on a rainbow.
Reply With Quote
  #108  
Old 02-14-2018, 01:38 PM
Colonel Smoothie's Avatar
Colonel Smoothie Colonel Smoothie is online now
Member
CAS
 
Join Date: Sep 2010
College: Jamba Juice University
Favorite beer: AO Amber Ale
Posts: 47,466
Default

Quote:
Originally Posted by whoanonstop View Post
In the end, I just don't have faith that spoon-fed programs will produce the highest tier people for the long term.
That's why I only hire people with no degree. You can't trust the ones that have been spoon-fed by attending school.
__________________
Recommended Readings for the EL Actuary || Recommended Readings for the EB Actuary

Quote:
Originally Posted by Wigmeister General View Post
Don't you even think about sending me your resume. I'll turn it into an origami boulder and return it to you.
Reply With Quote
  #109  
Old 02-14-2018, 07:33 PM
ActSciMan's Avatar
ActSciMan ActSciMan is offline
Member
CAS
 
Join Date: Nov 2010
Studying for data science
Posts: 1,859
Default

Quote:
Originally Posted by whoanonstop View Post
Noticed the same thing myself. Huge red flag as I have little faith the majority of those people could handle any technical rigor. Even better, note that they point out under education their "SAS credentials", which I'm pretty sure every candidate has. Must be a pre-requisite for the program?

http://analytics.ncsu.edu/reports/MS...t_Profiles.pdf
-Riley
I assume this means they teach a lot of SAS in their classes, possibly with some emphasis on obtaining these certifications. This would make sense given that SAS is headquartered very near there, they probably have a very strong relationship.
Reply With Quote
  #110  
Old 02-14-2018, 07:45 PM
ActSciMan's Avatar
ActSciMan ActSciMan is offline
Member
CAS
 
Join Date: Nov 2010
Studying for data science
Posts: 1,859
Default

My question, say someone wants to learn on their own, what are the most important topics to learn (or perhaps most important study materials)? Here are my specifics.

I'm currently working through An Introduction to Statistical Learning. I've started this in the past, but I will actually finish this time, reading the entire book, watching the corresponding lecture videos by two of the authors, doing the R labs, doing at least some of the exercises so I have some experience doing the modeling in R. I'm about halfway through at this point. I'm also reading The Art of R Programming, not quite half way through that at this point.

What's next? Books I think I first heard about on the AO include The Elements of Statistical Learning, Applied Predictive Modeling by Kuhn, Machine Learning: A Probabilistic Approach by Murphy. Any of these? Others? ESL has the advantage that it's free, but I'm fine paying for a good book.

For those who don't remember or know, I have a PhD in math, so I'm not worried about the math being too difficult.
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


All times are GMT -4. The time now is 12:32 PM.


Powered by vBulletin®
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
*PLEASE NOTE: Posts are not checked for accuracy, and do not
represent the views of the Actuarial Outpost or its sponsors.
Page generated in 0.39087 seconds with 11 queries