As of yesterday, a toll system is in place for drivers crossing the I-520 bridge, linking Seattle to Redmond, where the big Microsoft campus is located. It is claimed to be the most modern toll system in US, and drivers don't need to stop to cross the bridge, there's actually no toll booths, no human beings to process your payment.…

The information age is also the age of information overload. Companies, governments, researchers and private citizens are accumulating digital data at an unprecedented rate, and amid all those quintillions of bytes could be the answers to questions of vital human interest: What environmental conditions contribute most to disease outbreaks? What sociopolitical factors contribute most to educational success? What player statistics best predict a baseball team’s win-loss…

I lived in California a while back, and have experienced a few instances where large companies headquartered e.g. in Minneapolis, were selling products in all their stores nationwide as if everybody in US had the same weather, seasons and body size.

In particular, some stores (maybe Target - I don't remember) that sell garden tables did not have any product available in late February, although late February in much of California is a great time to remodel your garden.

Also,…

WK1model

In the words of Dr. Chris Chatfield: “Predictions are often given as point forecasts with no guidance as to their likely accuracy (and perhaps even with an unreasonable high number of significant digits implying spurious accuracy!)”.

Our paper sets forth a synergy of existing statistical theories to obtain a clear-cut model for calculating forecasts with prediction intervals, named the…

The message below is from Google, you can read the original version at www.google.com/explanation.html.

If you recently used Google to search for the word “Jew,” you may have seen results that were very disturbing. We assure you that the views expressed by the sites in your results are not in any way endorsed by Google. We’d like to explain why you’re seeing these results when you conduct this search.

A site’s…

Big data i s the new buzzword within the data warehousing and business analytics community.

If observations from a specific experiment (for instance, scores computed on 10 million credit card transactions) are assigned a random bin ID (labeled 1, ··· ,k), then you can easily build a confidence interval for any proportion or score computed on these k random bins, using the Analyticridge…

By improving its search and relevancy engines, to include item price as a main factor. The type of optimization and ROI boosting described below applies to all digital catalogs. Here we focus on books.…

The solution is simple: leverage external data, and simplify your predictive model.

Back in 2000 I was working with GE's analytic team to improve sales forecasts for NBC Internet, a web portal owned by NBC. The sales / finance…

The credit card transaction described here in details is a real example of a fraudulent transaction performed by organized criminals, undetected by all financial institutions involved, and very easy to detect with simple text mining techniques.

It was not caught by any of the financial…

Who would have guessed, five years ago, that by 2011 we would have mobile phones that would, in a rudimentary way, listen to our questions and give us useful answers? Or ways for doctors to get help to…

Hiring managers always assumed that I was a very detailed-oriented person. It turns out that this is not the case: I'm certainly a very analytic person, yet I always think "big picture", and everybody who knows me well would say that I am everything but detail-oriented.…

According to the recent landmark McKinsey Big Data study, by 2018 one in two US jobs requiring deep analytical skills will go unfilled, and there will also be an under-served demand for 1.5 million business managers and analysts “who can ask the right questions and consume the results of the analysis of big data effectively.”

McKinsey and others see it…

The strategy is very easy to describe:

1. Divide the domain of your function in k sub intervals.

2. Initialize k monomials;

3. Consider the monomials as centroids of your clustering algorithm.

4. Assign the points of the function to each monomial in compliance to the cluster algo.

5. Use the gradient descent to adjust the parameters of each monomial.

6. Go to 4. until the accuracy is good enough.

Read the entire post at:…

Need to prove something you already believe? Statistics are easy: All you need are two graphs and a leading question.

By Vali Chandrasekaran

Correlation may not imply causation, but it sure can help us insinuate it.

Source: …

Posted by Leonard Murphy.