Скачать с ютуб GMU/WSS Joint Seminar on the November Election в хорошем качестве

GMU/WSS Joint Seminar on the November Election

Please join the Washington Statistical Society (WSS) and George Mason University (GMU) for a joint seminar on the November election. Andrew Gelman (Professor of Statistics and Political Science, Columbia University) will be interviewed by Edward Wu (Data, Polling, and Election Analytics Senior Producer, CNN), followed by Dr. Gelman’s talk “Tough choices in election forecasting: All the things that can go wrong”. Biography - Andrew Gelman: Andrew Gelman is a professor of statistics and political science at Columbia University. His books include Bayesian Data Analysis, Regression and Other Stories, Active Statistics, and Red State, Blue State, Rich State, Poor State: Why Americans Vote the Way They Do. He and his colleagues developed multilevel regression and poststratification (MRP), which is used by YouGov and other survey organizations to obtain local estimates from imperfect samples. He has served on the board of the General Social Survey and has worked with The Economist magazine on their election forecasts in 2020 and 2024. Biography - Edward Wu: Edward Wu is Data, Polling, and Election Analytics Senior Producer, based in CNN’s Washington, D.C. bureau. Ed helps to lead CNN’s Election Night Decision Desk and works on the network’s election and polling methodologies. Prior to joining CNN, Ed held positions as research faculty in the University of Virginia Biocomplexity Institute and Science Policy Fellow at the American Statistical Association. He has a Ph.D. in Statistics from the University of Michigan and a bachelor’s degree in Mathematics-Statistics from Columbia University. Abstract for "Tough choices in election forecasting: All the things that can go wrong”: Different election forecasts use different models but yield similar predictions, which is no surprise as they're using the same information: state polls, national polls, and what we call the "fundamentals": economic conditions, presidential popularity, incumbency, and the relative positions of the states in previous elections. And when the forecast is close to 50/50, which it is at the time of this writing, it's hard to be called out. Back in the day, it was considered impressive for a forecast to predict 45 or more states correctly, but now everyone knows which are the key swing states, and it's not hard to come up with a reasonable state-by-state forecast with historically-calibrated uncertainty. Nonetheless, there are lots of ways a forecast can go wrong, and lots of choices involved in modeling, data inclusion, testing and validation of the forecasting procedure, and communication of results. We discuss this in the context of our efforts in election forecasts from 1992 to the present day.