PhD Thesis DefensesSpring 2017 Tuesday, March 21, 2017 Friday, March 24, 2017 |
MS Thesis DefensesSpring 2017 Wednesday, March 8, 2017 Wednesday, March 22, 2017 Friday, March 31, 2017 Friday, March 31, 2017 Friday, March 31, 2017 Wednesday, April 5, 2017 Wednesday, April 5, 2017 Thursday, April 27, 2017 |
Senior Capstone PresentationsSpring 2017 (Full itinerary here) Tuesday, May 2, 2017 Abstract:There are several factors playing a critical role in impacting real estate price of China’s large cities. These factors include GDP, population, immigration, mileage of Urban Rail Transit facilities, average saving of a person, and average consumption of a person. Using data on these factors and data of housing prices, this study constructs a multivariable linear model to evaluate the effect of these variables on housing quantitatively and graphically. For example, a correlation test reveals significant correlations between housing price and these factors and a scatterplot reveals the linear relationship between housing price and these factors. I also use extra sums of squares in Tests for regression Coefficients to see whether some coefficients can or cannot be dropped from the linear model.
Tuesday, May 2, 2017 Abstract: Election turnout varies over a wide range both year to year and from county to county in the state of Ohio. An analysis was run on each of Ohio’s 88 counties with eight different factors to discover their effect on election turnout over the last twenty years. These factors include previous turnout, unemployment, population density, minority population, income, education level, partisanship in the previous Presidential election, and the poverty level. Analysis was performed using multiple linear regression and principal component analysis using Excel to organize the data and R to run the analyses.
Thursday, May 4, 2017 Student: Isabelle Wagner Abstract: I’ll discuss the variables used in the analysis of their salaries and some basic visualizations showing how they relate to it. Next I’ll show the multiple regression model including all of the variables and analyze how good of a model this is for predicting salary. Then, I’ll discuss modifications made to the model to improve its accuracy in prediction. Finally, I’ll explain the conclusions I was able to come to from this project. Student: Thomas Nolan Abstract: The NCAA Division I Men’s Basketball Tournament is one of the most watched annual sporting events in the United States, and one important aspect of the event is the selection and seedings of the teams participating in the tournament. The goal of this study is to use statistical modeling to mimic the thought-process of the Selection Committee to accurately predict the 68 participants of the NCAA Tournament and their seeds. The best model for predicting at-large teams receiving bids is one that has the highest accuracy of correctly selecting the teams. The best model for determining seeding is one that minimizes the mean square error of the predicted seed compared to the actual seed. In this paper, I will give a more detailed background of the NCAA Tournament selection process, the data and variables that are likely to be used by the Selection Committee, and the modeling process and prediction results. Abstract: This presentation deals with the survival rates of breast cancer patients on various factors. This was achieved by obtaining from a German Breast Cancer study and assessing the data in order to create relevant models for the data. Model diagnostics and validation were carried out in order to determine the best model from assessment for the data. The resulting model details some key factors in the survival rates of breast cancer patients. |
For a list of past student presentations, please click here.