E46 BMW Social Directory E46 FAQ 3-Series Discussion Forums BMW Photo Gallery BMW 3-Series Technical Information E46 Fanatics - The Ultimate BMW Resource BMW Vendors General E46 Forum The Tire Rack's Tire Wheel Forum Forced Induction Forum The Off-Topic The E46 BMW Showroom For Sale, For Trade or Wanting to Buy

Welcome to the E46Fanatics forums. E46Fanatics is the premiere website for BMW 3 series owners around the world with interactive forums, a geographical enthusiast directory, photo galleries, and technical information for BMW enthusiasts.

You are currently viewing our boards as a guest which gives you limited access to view most discussions and access our other features. By joining our free community you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is fast, simple and absolutely free so please, join our community today!

If you have any problems with the registration process or your account login, please contact contact us.

Go Back   E46Fanatics > Everything Else > The Off-Topic > General Off-Topic

General Off-Topic
Everything not about BMWs. Posts must be "primetime" safe and in good taste. You must be logged in to see sub-forums.
Click here to browse all new posts.

Reply
 
Thread Tools Search this Thread Rate Thread Display Modes
Old 08-21-2014, 08:38 PM   #1
bimmerfan08
Registered User
 
Join Date: Feb 2009
Location: Hampton Roads, VA
Posts: 4,891
My Ride: Phoenix Yellow M3
MIT SDM Webinar "Move Over, Big Data! How Small, Simple Models Can Yield Big Insights

Quote:
MIT SDM Systems Thinking Webinar Series

September 8, 2014
Noon - 1 p.m. Eastern time
Free and open to all

Details/registration

Today's emphasis on big data and data analytics may leave some folks thinking that management and policy insights can only arise from the analysis of millions of data entries. Nothing could be further from the truth! Sometimes less is more. In fact, an excess of numbers can engender more headaches than insight.

In this talk, managers and policymakers will learn how simple mathematical models of systems can improve intuition and lead to better decisions. Dr. Larson will provide concrete examples from his professional research and consulting engagements, then discuss general applications to industry. He will cover:

-Flaws of averages-what they are and how to avoid them;
-Square root laws-how to apply them to locating facilities and more;
-Singularities-why and how managers of service systems must schedule idle time for servers or face huge waiting lines (aka the "elbow effect");
-Simple difference equations-how to use them to discover major system instabilities when inputs are year-to-year gross revenues;
-Going viral-how a major demography parameter can apply to exponential explosiveness in many business sectors; and
-Lateral thinking-and how it can sometimes make a problem go away.

Learn to cut to the chase, see the big picture, and stay out of the weeds!

A Q&A will follow the presentation. We invite you to join us.

About the Speaker

Professor Richard C. Larson has been a member of the MIT faculty for more than four decades, in four different academic departments. During this time he has also led an off-campus consulting firm that has invented novel approaches-inspired from operations research and industrial engineering-to complex systems problems in the private and public sectors. He has served as president of both the Operations Research Society of America and the Institute for Operations Research and the Management Sciences. He has worked closely with a wide variety of organizations, including-in the private sector-banks, airlines, retailers, industrial gas distributors, amusement parks, and-in the public sector-the City of New York, many public school systems, the U.S. Postal Service, the World Bank, the Centers for Disease Control, the National Institutes of Health, and numerous police departments. At MIT he has founded several initiatives, including MIT Learning International Networks Consortium and MIT Blended Learning Open Source Science or Math Studies.

Dr. Larson is a member of the National Academy of Engineering (NAE) and cochairs a major panel on the application of systems engineering to health, cosponsored by the NAE and the Institute of Medicine.
Registration is free and open here: http://sdm.mit.edu/news/news_article...ll-models.html
__________________
Quote:
The whole business of politics has been effectively subcontracted out to a band of professionals. Money people, outreach people, message people, research people. The rest of us are meant to feel like amateurs. In the sense of suckers. We become demotivated to learn more about how things work. We begin to opt out.

Last edited by bimmerfan08; 08-21-2014 at 08:41 PM.
bimmerfan08 is offline   Reply With Quote
Old 08-23-2014, 12:19 PM   #2
325kev
Registered User
 
Join Date: Feb 2011
Location: Tempe, AZ
Posts: 69
My Ride: 03 325i 5MT
In on nerd thread!

What he's referring to (small models) are basically queuing and other predictive analytic models. To me, they serve different purposes than "big data". Where modeling is used to predict system behaviors (what happens to my resources or customers as demand grows), big data is used to leverage information and target a customer base (people who like product A and also likely to buy product B). Big data can help identify that and then give information to target specifically that population.

Seems like he's just using the "Big Data" tag to catch people's attention as it is a hot topic right now. Both are useful in their own ways.
325kev is offline   Reply With Quote
Old 08-23-2014, 12:31 PM   #3
bimmerfan08
Registered User
 
Join Date: Feb 2009
Location: Hampton Roads, VA
Posts: 4,891
My Ride: Phoenix Yellow M3
I'm surprised Zell hasn't stopped in yet.
__________________
Quote:
The whole business of politics has been effectively subcontracted out to a band of professionals. Money people, outreach people, message people, research people. The rest of us are meant to feel like amateurs. In the sense of suckers. We become demotivated to learn more about how things work. We begin to opt out.
bimmerfan08 is offline   Reply With Quote
Old 08-23-2014, 01:17 PM   #4
Act of God
Registered User
 
Join Date: Dec 2004
Location: NY
Posts: 334
My Ride: 3.5 Liters of fury
Send a message via AIM to Act of God
Quote:
Originally Posted by bimmerfan08 View Post
I'm surprised Zell hasn't stopped in yet.
he's preparing the noose
__________________
Gold Medal Recipient: Jimmy Rustling (2014)

“They have the guns and therefore we are for peace and for reformation through the ballot. When we have the guns then it will be through the bullet.” - Saul Alinsky, quoting Lenin
Act of God is offline   Reply With Quote
Old 08-23-2014, 02:33 PM   #5
Zell
Registered User
 
Zell's Avatar
 
Join Date: Jun 2009
Location: Such City
Posts: 5,683
My Ride: '02 Dogemobile Shibe
Quote:
Originally Posted by bimmerfan08 View Post
I'm surprised Zell hasn't stopped in yet.
Weekend brah!

Some of my most effective models have come from "small data." When you account for probabilities of groups being selected, it's possible to create some fantastic insights. I solved a few probability problems and applied the inverse of them as weights to inputs into some penalized logistic regression models. We ended up gaining the best insight out of all other models.

There are two types of modeling one can do nowadays: explanatory, and predictive. Explanatory you try to find actionable items; that is, things you can affect as a company. Predictive models sometimes have no good explanation, but in the end, "Who cares? It works." I've found some strange and weird predictive models using statistically-sound variables that seemingly didn't make sense, but were demonstrably showing predictive power. Big data allows us to directly test that by holding out certain parts of data for testing predictions, and models can be tuned for that. Why is one variable good for prediction? Well, occasionally we just don't exactly know. We can guess all day and get close to an answer, but it doesn't change anything about the model. Data transformations, squaring things, etc. can all give good predictive power, but it might not be useful for explanation. If it's working on the hold-out data, might as well give it a shot. 90-95% of the time, it will work (literally).

I generally keep it simple when I'm trying to explain something, but go balls-out complex when trying to predict something. In a predictive model, no one will be asking why something is related to the dependent variable. They just want to know if we can predict it better.
__________________

Last edited by Zell; 08-23-2014 at 02:40 PM.
Zell is online now   Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes Rate This Thread
Rate This Thread:

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 On
Censor is ON





All times are GMT -5. The time now is 02:09 PM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2014, vBulletin Solutions, Inc.
(c) 1999 - 2011 performanceIX Inc - privacy policy - terms of use