Duquesne Undergraduate Research in Math and Computer Science (advisor: John Kern)


Jeromy Sivek. Presented at the 2007 MAA Regional Meetings at Mercyhurst College in Erie, PA, April 13, 2007. Also presented at Duquesne University Undergraduate Research Day, April 17, 2007.

Title: EVALUATING THE VALIDITY OF NON-EUCLIDEAN DISTANCE METRICS IN GAUSSIAN PROCESS MODELS.

Overview: Geodesic distance, an intuitive and desirable metric in certain spatial statistics applications, can yield invalid covariograms when the covariogram parameters are outside of a certain set. Identify the subset of the parameter space that yields valid covariograms, and constrain parameter estimation to that set.

Link to presentation slides:

Presentation [pdf]



Aubrey Komorowski. Presented at the MathFest 2005 national meetings in Albuquerque, NM, August 6, 2005. Also presented at the University of Nebraska, and at Juniata College.

Title: INDEPENDENCE MODELS, LIKELIHOOD RATIO TESTS, AND A SIDE OF BACON.

Overview: The game PASS THE PIGS contains two small rubber pigs. These pigs are used as dice, and the positions of the rolled pigs translate directly to points for (or against) the player who rolled them. Does the height from which the pigs are rolled (5 inches vs. 8 inches) significantly impact the scoring probabilities?

Link to presentation slides:

Presentation [pdf]



Elizabeth Ann (BA) Tiedeman. Presented at the MathFest 2005 national meetings in Albuquerque, NM, August 6, 2005.

Title: a PRICE PREDICTION MODEL FOR BUILDING BLOCKS.

Overview: The price of a LEGO set is certainly a function of, among other things, the number of pieces in the set. What kind of function, though? And, what other variables are significant predictors of the price of a LEGO set?

Link to presentation slides:

Presentation [pdf]



Neda Khalili and Janeen Peretin. Presented (jointly by Neda and Janeen) at the MathFest 2001 national meetings in Madison, WI, August 2, 2001.

Title: FROM MATHEMATICS TO KRYPTON: IN PURSUIT OF RANDOM NUMBERS.

Overview: Establish the usefulness of random number generation. Survey computer random number generation algorithms. Identify randomness tests for a list of (assumed) random numbers. Recognize Krypton as a random number generating tool that does not follow any algorithm.

Link to presentation slides:

Presentation [pdf]

Duquesne Home | Computational Mathematics Home | Kern Home