Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Property - Casualty / General Insurance
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions


Fill in a brief DW Simpson Registration Form
to be contacted when new jobs meet your criteria.


Reply
 
Thread Tools Search this Thread Display Modes
  #21  
Old 06-18-2019, 09:19 AM
Actuarially Me Actuarially Me is offline
Member
CAS
 
Join Date: Jun 2013
Posts: 191
Default

Quote:
Originally Posted by Lucy View Post
I know what predictive modeling and automated reporting mean in this context, but I'm not sure what you mean by data engineering or "quality of life improvements".

(My typical quality of life improvement is buying a new chair...)
Data engineering is building databases and data processes. Our SQL server is a mess, so someone has to dig in and scrub the data so it's easily accessible and in formats that people without SQL technical skills can access it. For instance, you can create a Shiny app, to create a UI that does all the SQL querying for people that spits out in a nice .csv at the click of the button.

It also how you deploy models and calculations. Hosting the models on a server, turning them into an API and ensuring all of the proper proxies are set. Sometimes that's the job of the IT department, but in my case it's not.

So data engineers are more on the programming side of things (where knowing python comes in handy) and data scientists are more on the statistical side of things.

quality of life improvements is having my boss' coffee ready by 8:45, so it's just the right temperature by the time they arrive.

In my case, an example of QoL improvements are creating scripts that read and parse PDF's so coworkers don't have to spend hours getting them in a usable format. I get a list of tedious tasks from a data perspective and find ways to make it less tedious. I enjoy that aspect the more than building models sometimes because I remember doing a lot of annoying manual tasks when I was in actuarial consulting.
Reply With Quote
  #22  
Old 06-18-2019, 09:24 AM
Lucy's Avatar
Lucy Lucy is offline
Member
 
Join Date: Sep 2001
Posts: 2,768
Default

Thanks. That all makes sense.
Reply With Quote
  #23  
Old 06-18-2019, 11:06 AM
bjc2142's Avatar
bjc2142 bjc2142 is offline
Member
CAS
 
Join Date: Aug 2014
Location: Boston
Studying for Python
Favorite beer: Asahi Draft
Posts: 1,193
Default

Seems like it’s very different across companies even more so than I expected. Here we have large pool of data scientists. I went to internal Kaggle-ish competition info session and had a feeling Python was preferred over R here. Competition related trainings were in Python and examples were shown in Python. R users could still participate. I guess R vs Python usage at work depends on proportion of non-actuarial data scientists : actuaries. A lot of masters program in DS utilize Python these days, so recent DS grads might be more used to Python than to R.

Do you know if people use Actuar R package? One thing I wanted to do to master Python while brushing up basic R skill was to rewrite actuar package in Python (for learning purpose initially). If it’s not really used, maybe there is less value in doing so. For work, I used Meyer’s MCMC code which may depend on that or CL package but I’m not too sure…
__________________
FCAS - Hawaii 2019
Reply With Quote
  #24  
Old 06-18-2019, 11:34 AM
Heywood J Heywood J is offline
Member
CAS
 
Join Date: Jun 2006
Posts: 4,093
Default

I think R vs. Python debate is really a debate between statistically-minded data scientists and CS-minded data scientists. If you're dealing with a flood of data, and your challenge is to make effective use of it, then Python would probably be the language of choice for you. If you're trying to tease out signal from your noisy dataset, then R would be the more appropriate language.

If you work for Facebook, and your goal is to figure out how to use gazillion terabytes of data to make people waste more time on Facebook, then you'll be more effective with Python. For typical actuarial uses of data science, I really think that the use of Python should be discouraged. Your choice of tools will bias you towards the kinds of models that you build, and Python will steer you towards models that are ill-suited for insurance.
Reply With Quote
  #25  
Old 06-18-2019, 01:26 PM
Vorian Atreides's Avatar
Vorian Atreides Vorian Atreides is offline
Wiki/Note Contributor
CAS
 
Join Date: Apr 2005
Location: As far as 3 cups of sugar will take you
Studying for ACAS
College: Hard Knocks
Favorite beer: Most German dark lagers
Posts: 66,137
Default

Quote:
Originally Posted by Heywood J View Post
I think R vs. Python debate is really a debate between statistically-minded data scientists and CS-minded data scientists. If you're dealing with a flood of data, and your challenge is to make effective use of it, then Python would probably be the language of choice for you. If you're trying to tease out signal from your noisy dataset, then R would be the more appropriate language.

If you work for Facebook, and your goal is to figure out how to use gazillion terabytes of data to make people waste more time on Facebook, then you'll be more effective with Python. For typical actuarial uses of data science, I really think that the use of Python should be discouraged. Your choice of tools will bias you towards the kinds of models that you build, and Python will steer you towards models that are ill-suited for insurance.
I would also imagine that UI requirements along with "frequency of use/development" could also be a consideration between Python and R (i.e., R-shiny).
__________________
I find your lack of faith disturbing

Why should I worry about dying? Itís not going to happen in my lifetime!


Freedom of speech is not a license to discourtesy

#BLACKMATTERLIVES
Reply With Quote
  #26  
Old 06-18-2019, 01:28 PM
Vorian Atreides's Avatar
Vorian Atreides Vorian Atreides is offline
Wiki/Note Contributor
CAS
 
Join Date: Apr 2005
Location: As far as 3 cups of sugar will take you
Studying for ACAS
College: Hard Knocks
Favorite beer: Most German dark lagers
Posts: 66,137
Default

Also, is there anything analogous to R-Studio for Python?

I see (with a very quick search) that there are some functionality for working with Python within Visual Studio; but I don't think that this would be the same sort of thing as you get between R and R-Studio.
__________________
I find your lack of faith disturbing

Why should I worry about dying? Itís not going to happen in my lifetime!


Freedom of speech is not a license to discourtesy

#BLACKMATTERLIVES
Reply With Quote
  #27  
Old 06-18-2019, 02:06 PM
Heywood J Heywood J is offline
Member
CAS
 
Join Date: Jun 2006
Posts: 4,093
Default

There is nothing that comes close to R Studio in Python, as far as I'm aware. The ease with which R Studio allows you to inspect intermediate output on demand is another reason why actuaries should opt for R over Python. The standards for error checking and common sense checking are much higher for actuaries than for typical data scientists.
Reply With Quote
  #28  
Old 06-18-2019, 02:08 PM
bjc2142's Avatar
bjc2142 bjc2142 is offline
Member
CAS
 
Join Date: Aug 2014
Location: Boston
Studying for Python
Favorite beer: Asahi Draft
Posts: 1,193
Default

Quote:
Originally Posted by Vorian Atreides View Post
Also, is there anything analogous to R-Studio for Python?

I see (with a very quick search) that there are some functionality for working with Python within Visual Studio; but I don't think that this would be the same sort of thing as you get between R and R-Studio.
DS is usually done in Jupyter Notebook. That's analogous to Rmarkdown. It's very easy to use. Programming is usually done in a text editor based on personal preference.
__________________
FCAS - Hawaii 2019
Reply With Quote
  #29  
Old 06-18-2019, 02:09 PM
bjc2142's Avatar
bjc2142 bjc2142 is offline
Member
CAS
 
Join Date: Aug 2014
Location: Boston
Studying for Python
Favorite beer: Asahi Draft
Posts: 1,193
Default

Quote:
Originally Posted by Heywood J View Post
The ease with which R Studio allows you to inspect intermediate output on demand is another reason why actuaries should opt for R over Python.
That's why Jupyter Notebook is used for DS work in Python.
__________________
FCAS - Hawaii 2019
Reply With Quote
  #30  
Old 06-19-2019, 01:56 AM
Avi's Avatar
Avi Avi is offline
Wiki Contributor
Site Supporter
Site Supporter
CAS AAA
 
Join Date: Aug 2002
Location: NY
Studying for the rest of my life.
College: Alumnus - Queens College - CUNY
Favorite beer: Stone Ruination IPA
Posts: 14,212
Blog Entries: 3
Default

Quote:
Originally Posted by Vorian Atreides View Post
Also, is there anything analogous to R-Studio for Python?

I see (with a very quick search) that there are some functionality for working with Python within Visual Studio; but I don't think that this would be the same sort of thing as you get between R and R-Studio.
Not really, but I use JetBrains PyCharm as my Python IDE. Then there is Apache Zeppelin for on-prem Spark, and I really have to start spending some time with DataBricks.
__________________
All scientists defer only to physicists
Physicists defer only to mathematicians
Mathematicians defer only to G-d!

--with apologies to Dr. Leon Lederman
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 01:42 PM.


Powered by vBulletin®
Copyright ©2000 - 2019, 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 0.49706 seconds with 9 queries