Student Research Page

This is a page dedicated to Student Projects in the undergraduate Statistics class MA331. Some of these projects are very good and I have decided to publish the best of them every year. For more detail and quesrtions about the projects please contact the authors. If details are not printed within the article a search for the author's name here is likely to produce results. In the later year the projects are improving because the authors are required to have more structure in their writing.

Projects from 2010

Ian Cordasco, Alaina Spicer, Tadas Vilkeliskis, and Robert Williams What Causes Violent Crime?
Abstract: The amount of violent crime in a community affects all aspects of life in that community – the economy, the mood of residents, the careers of the politicians in power, etc. Unfortunately, violent crime is a hard variable to measure because many such crimes go unreported. The goal of this study is to find a statistically significant model that can predict the violent crime in a community using explanatory variables that are readily available in public data sets. In doing this, the resulting model will allow law makers and police to understand if their communities exhibit unexpectedly high or low crime – this could be used to prompt an investigation and get a better understanding of crime in that community. Through linear and logistic regression, two useful models were found that can be used for these purposes. Additional experimental models such as neural networks were investigated with lesser success. Article and Presentation TRSTAT.2010.1

Danielle Maginnis, Justin Sousa, and Tom Zinckgraf. DEA $$$ Spending; Necessary?
Abstract: The purpose of our project was to determine the percentage of the Drug Enforcement Administration’s (DEA) budget that is being spent on drug enforcement is unnecessary. Basically, when looking at the available data on drug usage, we will prove that the constant increase in the Drug Enforcement Administration’s budget is not making a positive impact for the Drug Enforcement Administration. Article and Presentation, TRSTAT.2010.2

Abhijit Amin, Jamison Andaluz, Robin Azzam, Nirmal Rajan, and Richard Soni: Cars Going Green
Abstract: Carbon Footprint is a hot topic in our current society. Cars play a key role in Greenhouse Gas emissions. If the amount of toxins being launched into our atmosphere does not diminish our children will suffer. We attempted to find the elements in a motor vehicle that affects its GHGS (defined below). We found that there are variables that affect this, and that some can be used to predict the GHGS of a vehicle. The model that best explains the significant effect on the Greenhouse Gas Score of a vehicle, and the one we used for our predictions, uses its combined mileage, engine displacement, SmartWay Certification, air pollution score(APS) and drive type (GHGS ~ Cmb + Displ + SW + APS + Drive). Further explanation of these variables can be found in the definition of variables. An interesting finding in searching for our model was that dependent on the mileage type, combined; highway; or city, the factors affecting the Greenhouse Gas Score differ. Our best model used the vehicle’s combined mileage. Another interesting finding is the strength of the air pollution score (APS). The group assumed that APS would be one of the most significant variables in the GHGS. However, in our model, the APS was less significant than combined mileage and engine displacement. Article and Presentation TRSTAT.2010.3

Daniel Bolella, Brian Donohue, David Fonorow, and Benjamin Rose: Predicting Champions: Using season stats to predict sports champions!
Abstract: Post-season performance by a sports team is largely dependent on circumstances surrounding the game currently being played. However, it seems that some sports teams continually outperform others on a year by year basis. Therefore circumstantial variables are not the only factors that indicate playoff performance. Considering there is a wealth of data over past sports seasons, it is possible to perform an analysis on the data available in an attempt to determine playoff performance. The outcomes provided by such a study can help coaches, players, and fans alike in determining what team qualities are the most beneficial for playoff performance in a given sport. Article and Presentation TRSTAT.2010.4

Projects from 2009

Jeffrey Cochran, Jason VanBuiten, Michael Smith Exploring Character Recognition through Modeling: Cognitive Reverse Engineering
Abstract: What makes letters harder or easier to read? Moreover, what factors are most important to the brain for successful character recognition? These were the concepts investigated in this study. A “critical point theory” was developed that suggested that the brain uses intersection points and points of inflection in letters for pattern recognition. An experiment was developed in which letters were shown to subjects that had certain factors controlled such as occlusion of the letter, skew, and frequency of the letter in English. Data was collected as the time it took for the subjects to name each letter. A multiple linear regression model was developed, analyzed, and refined based on this data. The model was intended to be used to recognize what factors were significantly important for character recognition, and to predict response time of hypothetical letters. Statistical evidence of the validity of the critical point theory was found. However, the study suffered from poor experimental design, and the resulting regression model was unsatisfactory for use in prediction. A follow-up study with a revised experiment is suggested, and our results are discussed. Article and Presentation TRSTAT.1.2009

Mengting Guo, Yuxiao Ning, Haig Shishmanian Investing in Major Film Productions: Analyzing Profitability Determined by Film Characteristics
The film industry continually produces very successful box-office hits. Where there is profit, there is investment opportunity. A study was done for the purpose of advising potential investors in what films to become financiers of. Specific aspects of films were chosen as factors potentially contributing to profitability. Recent films from the past 5 years were sorted based on 3 factors: The source of the story, the genre of the film, and the MPAA rating given to the film.
First, the analysis showed that there is no direct relationship between budget size and profitability. Next, two very profitable combinations were determined: PG-rated Adventure Sequels and PG-rated Dramas based on original stories. Lastly, G-rated films were analyzed separately because of their being rare and consistently profitable.Article and Presentation TRSTAT.2.2009

Danielle Bisordi and Kristin Buckley Statistical Analysis of Near Earth Objects
In this paper, we explore the statistical properties of Near Earth Ob jects (NEOs). It is determined whether the date has an effect on the quantity of NEOs observed, as well as if the velocities, sizes and orbit types of the NEO are correlated with the miss distances. Article and Presentation TRSTAT.3.2009

Sean McCoy, Robert Hoffmann, Vincent Lipoma, and Sean Mullaney: Regulation and Gun Homicide in the U.S.
This study focuses on the effects of regulation on gun homicide in the United States. We gathered data on homicides and gun regulations in different states and performed tests to see whether gun regulation, in the form of mandatory permits and registration, had an effect on gun homicide rates. We obtained mixed results about the significance of effects of regulation on the gun homicide rate. In addition, we sought to produce a model for predicting gun homicide rates based on regulation as well as other influential factors. Article and Presentation TRSTAT.4.2009

Projects from 2008

Jamie Brabston, Matt Caulfield and Mark Testa The Wealth of Nations
We did a study that attempted to identify some of the economic, governmental, and demographic factors that affect or are affected by how wealthy a country is. We are looking at thirty countries from around the world, selected for diversity in culture, location, and economic status. For each country, we have obtained from the CIA World Factbook economic data such as national debt, oil imports and exports, and unemployment rate, and demographic data such as life expectancy, population density, and literacy rate. We performed a multivariate regression analysis to determine the correlation between these and other variables, and GDP per capita. Once we had identified significant variables and their relationship to GDP, we attempted to identify the causal relationship between the variable and GDP: whether one causes the other, or if both are caused by a lurking variable. We hoped to identify relationships between economics, government, and demographics that are not immediately obvious. Article and Presentation TRSTAT.1.2008

Christy King Internet Book vendors: Are they different?
The general objective of this study is to determine the significance in vendor selection when purchasing Fantasy books online. Article 1,2 TRSTAT.2.2008

Ryan Oelkers, Michael Hummel: Pennsylvania Voters: Party Affiliation Factors
The objective of this work is to determine the effect of location or age on the political affiliation of Pennsylvania voters Article TRSTAT.3.2008

Brittany Fuller and Ian Parks: A basketball study.
Does an NCAA women’s basketball team have a home court advantage? Article TRSTAT.4.2008

Projects from 2007

Krishna Hajari, Faraz Hyder, William Walker. Baseball statistics
Our goal is to find out if, over the past 10 years, there is a consistent factor that affects the winning percentage of the 30 teams in the Major League Baseball. Article and Presentation TRSTAT.1.2007

Giovanni Gaccione, Joe Trinsey, Thomas Zygnerski: Wired vs. Wireless Internet Usage on Steven Campus
A study was performed at the Stevens Institute of Technology which compared internet usage among various categories. The goal of the study was threefold: to determine whether wired or wireless internet was used more by students, to determine whether certain days of the week had greater traffic, specifically whether there was greater traffic on weekends, and to determine whether there was greater traffic at certain times of the day, specifically whether there was greater traffic in the afternoon and night compared to the morning hours. The study revealed interesting results. It was shown that for all dorms, wireless internet was used less than wired internet. No significant difference was detected in internet usage among days of the week. Finally, it was shown that there was significantly less internet usage in the morning than in the afternoon or nighttime. Article and Presentation TRSTAT..2007

Michael Cannito, Ronen Peled, Christina Sylvester: Wars in History - A look at the politics behind the Wars of the past
Is the GDP affected by the wars. Are the wars necessary for the economy. (Note: This one lacks an abstract so I had to describe it somehow. IF) Article and Presentation TRSTAT.3.2007

Rich Miktus, Christopher Geigel, Brandon Butch: Education in New Jersey (Again sorry for the lack of an abstract. IF) Article and Presentation TRSTAT.4.2007

Natalie Arndt, Allison Mucha: Determining Factors of GPA
Among full-time undergraduate Stevens students, number of credits and which school a student belongs to were the most important factors in determining GPA. Article and Presentation TRSTAT.5.2007