There are several paragraphs of text here, but the “tldr” is that although many systems increase in value dramatically in a superlinear manner with increasing scale, such as when there are economies of scale or network effects, we should be mindful that typically the value of data scales in a sublinear, logarithmic way. If … Continue reading The Unit Economics and Scaling of Data
There is a nice analysis at FiveThirtyEight by Nate Silver which looked at how Michelin reviews restaurants compared to how Yelp does. They were able to look at the restaurants which have or had a Michelin star in the past few years. Do Yelp reviews predict whether a restaurant will lose a star? Which restaurant … Continue reading Yelp Reviews vs Michelin Reviews
There is a great blog post over on http://beckmw.wordpress.com/2013/04/15/how-long-is-the-average-dissertation/ looking at dissertation length. It is definitely a standard based upon discipline, as we can see from this figure. There is some great R code over there to show how to analyze the data from the U of Minnesota's digital storage of dissertations and create the … Continue reading How long is an average PhD thesis?
If you are into statistical computing (or want to be), this is a great collection of freely available PDF texts on a range of topics. http://blog.revolutionanalytics.com/2014/03/an-r-meta-book.html _________________________________________________________________
I found this great post from Thomas Levine: http://thomaslevine.com/!/r-spells-for-data-wizards/ In it, he goes through some helpful tips and tricks for a bunch of common situations in R where things are a bit nonintuitive to a new user. It's really targeted for new-intermediate users. When you're ready to be more hardcore, you can move on … Continue reading R spells for data wizards
There is an excellent overview of Singular Value Decomposition (SVD) by David Austin here: http://www.ams.org/samplings/feature-column/fcarc-svd It provides a lot of good geometric intuition with nice figures and clear text.
Excellent overview on tracking down errors in R code from Pete Werner. http://petewerner.blogspot.com/2013/01/tracking-down-errors-in-r.html Shamelessly copying his post content below without permission: TUESDAY, JANUARY 29, 2013 Tracking down err...