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

Thread Tools Search this Thread Display Modes
Old 05-10-2019, 10:43 AM
Tacoactuary's Avatar
Tacoactuary Tacoactuary is offline
Join Date: Nov 2014
Location: Des Moines, IA
College: Vanderbilt, UIUC
Favorite beer: Yazoo Sue
Posts: 1,591

I'm looking for feedback and recommendations from others who have used or trialed use of h2o for predictive modelling work. I'm curious what your experiences have been with it and if you have found any limitations or had struggles using it in lieu of alternative modelling platforms.

In particular, I am interested in using it as an alternative/supplement to tools like Emblem.
Reply With Quote
Old 05-10-2019, 11:17 AM
Colonel Smoothie's Avatar
Colonel Smoothie Colonel Smoothie is online now
Join Date: Sep 2010
College: Jamba Juice University
Favorite beer: AO Amber Ale
Posts: 49,190

I like it. I've used both the R and Python versions, and appreciate the clean syntax and good documentation.
Recommended Readings for the EL Actuary || Recommended Readings for the EB Actuary

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
Old 05-28-2019, 11:05 AM
Starscream's Avatar
Starscream Starscream is offline
Join Date: May 2008
Location: Cybertron
Studying for 7 for the millionth time.
College: Eastern Michigan
Favorite beer: Samuel Smiths Taddy Porter
Posts: 77

I've used it for Loss Cost modeling. Trying to build demand models with it, w/ less success.

I like it. Data wrangling can be a bit annoying, so I use tidy packages and get as much done as i can outside of h2o, before converting to an h2o dataframe.

The flow web interface is nice for keeping tabs on the progress of your models.

The biggest problem I've found is that there are few insurance related examples, but that can be said about ML in general. Seems like "data scientists" focus on classification, or imagine recognition. Highly skewed insurance loss cost data just doesn't seem all that prevalent in the literature.
Real songs are about deals with the devil, far off lands, and where you find smoke in relation to water
Reply With Quote

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 08:21 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 0.14410 seconds with 11 queries