Sharing analysis

Here are three links that you should go through this week:

  1. If you’re look­ing for ways to share your ana­lys­is, RStudio 1.0 is out. The biggest fea­ture is R Notebooks, which are like Jupyter note­books. At Gramener, we’re us­ing RStudio server to col­lab­or­ate. Airbnb’s Knowledge Repo is an­other op­tion.
  2. If you’re fil­ter­ing data, be aware of Simpson’s para­dox. It ex­plains how Derek Jeter’s bat­ting av­er­age is higher than David Justice’s though the lat­ter per­formed bet­ter every year.
  3. Prepare for data sci­ence in­ter­views with this com­pil­a­tion of 109 data sci­ence in­ter­view ques­tions.

Speaking of Simpson’s para­dox, be wary of stat­ist­ic­al sig­ni­fic­ance as well:

One thought on “Sharing analysis”

Leave a Reply