Election Predictions

Data Skeptic

b'## Election Prediction\r\n\r\nINTRO VOICE-OVER: Data Skeptic features interviews with experts on topics related to data science, all through the eye of scientific skepticism.\r\n\r\nHOST: Jo Hardin has a PhD in statistics from UC Davis and is presently a Professor of Mathematics at Pomona College. Her work contributes to diverse topics such as computational biology and genomics. She recently wrote a blog post about the prediction competition for the US 2016 presidential election, giving some tips for competing, and with the election coming up, I thought it would be a great time to talk about how data scientists can study the election process.\r\n\r\nJo, welcome to Data Skeptic.\r\n\r\nJO: Thanks! Great to be here.\r\n\r\nHOST: Before we jump into the details and election discussion, could you tell us what the ASA is and what your particular association is to it?\r\n\r\nJO: The ASA is the American Statistical Association and I\x92m involved \x96 I have a couple of roles. I work closely with the statistics education section and their outreach to students of all different kinds. I also work locally with my chapter in Southern California.\r\n\r\nHOST: What can you tell me about the ASA\x92s Prediction Competition?\r\n\r\nJO: I think it\x92s just a way that the ASA came up with to really get students excited and involved with how you can use statistics to think about what\x92s going on in the news and really kind of current events. The prediction competition is pretty straightforward. It\x92s simply to predict both the winning presidential candidate as well as the final percentages of the popular vote for each major candidate and I think they want it also \x96 if you can also do it per state and then broken down by demographic groups and what not.\r\n\r\nThey want a thoughtful answer. They\x92ve asked for a 200 to 300-word description of the methods and what you did that was original. They\x92re not really looking for you to just report what Nate Silver has predicted.\r\n\r\nBut it\x92s kind of a fun thing and you c

Next Episodes

Data Skeptic

F1 Score @ Data Skeptic

📆 2016-09-23 02:00


Data Skeptic

Urban Congestion @ Data Skeptic

📆 2016-09-16 02:00


Data Skeptic

Urban Congestion @ Data Skeptic

📆 2016-09-16 02:00


Data Skeptic

Heteroskedasticity @ Data Skeptic

📆 2016-09-09 02:00


Data Skeptic

Front End and API Overview @ Data Skeptic

📆 2016-09-02 02:00