Overfitting: your forecasts may not be as good as the measure tells you

Job #12730

Job Posting Details

Job # 12730 Overfitting: your forecasts may not be as good as the measure tells you

Posted Date
Apr 20, 2009 @ 17:06
Respond By
Apr 22, 2009
Word Count
English (North American)
Age Range
Middle Aged

Job Description

We want to make a slidecast out of a gentle introduction to a statistical problem known as overfitting.

The scripts have been attached to this post. In order to make sense of the scripts, I suggest to have a look at the corresponding slides
http://joannes.vermorel.googlepages.com/Overfitting.pptx  (Powerpoint viewer may need to be installed to view)

The objective is to have a clean and very understandable voice for this job.


Note from Voices.com:  All attached documents as well as the link were created in Office 2007.  Voices.com has been able to review the script and the URL.

Sample from attached file:

Overfitting: your forecasts may not be as good as the measure tells you

Forecasting accuracy is critical for many industries such as retail, manufacturing or services. If you over-forecast your customer demand, your costs explodes because you will have too much staff, too much inventory. But if you under-forecast your customer demand, your customers get mad because they can’t buy your product or because they have to wait for too long to get served.

In this slidecast, I am going to introduce a too little known problem in forecasting called “overfitting”.

This problem is too little known for two reasons. First, it’s a subtle problem - non-obvious and counter-intuitive in many aspects. Second, it’s a problem that has been puzzling mathematicians since the 19th century. It’s only at the end of the nineties, a little more than 10 years ago, that the scientific community started to really comprehend this problem both at the theoretical level but also at the practical level.

Before getting any further, let me jump to the conclusion. Overfitting has a very strong impact on your forecasts. Overfitting can make you believe that you have a 10% forecast error while your real forecast error is 20%, and that would not be a worse case situation.

Ready to get started?

Join the #1 marketplace for voice-over talent

Need Some Help?
Contact us for help with your voice over project.