Contains methods for estimating covariance and precision matrices under the constraint of sparisty---i.e. that most of the off-diagonal entries are zero.
Student Contributors: Xinyu Zhang
Dataset comprised of total word counts and number of America's in each US presidents' state of the union and inaugural address.
This dataset was produced from the corpus of speeches and writings available thanks to the American Presidency Project
Some R code for analyzing this data using logistic and Poisson regression.
Dataset with 72 entries referring to Academy Awards broadcasts from 1942 to 2017 containing wordcounts for Best Picture, Leading Actor, and Leading Actress speeches as well as total budget, number of "thanks" in the speech, and total broadcast runtime.
Dataset with 24 entries recording the total wordcounts for 24 different Oscar speeches in both 2016 and 2017 with the goal of determining whether or not Jimmy Kimmel's offer of a jet ski for the shortest speech had any statistically significant effect on the speech lengths.
These dataset was produced from the corpus of speeches available thanks to the Academy Awards Acceptance Speech Database as well as Wikipedia for the runtimes, IMDB for the total budget, and the US Bureau of Labor Statistics for the historic inflation rate.