Embedded Files
Machine Learning Based Playstyle Classiffcation for NCAA Tennis Players
We present a machine learning framework for classifying NCAA tennis player playstyles using match data. Due to limited school-level data, models were trained on a professional dataset and applied to UCLA players through transfer learning. Gaussian Mixture Models generated soft labels for four playstyles, and a Random Forest Regressor achieved the best prediction performance. Playstyle scores were scaled using Universal Tennis Rating (UTR) for consistency across levels. Results were visualized via radar plots. This approach offers a scalable method for player analysis and highlights future improvements in data collection, feature expansion, and playstyle definitions.
Quantitative Financial Modeling (Issue I)
In the realm of financial mathematics, models play an indispensable role in capturing the inherent complexities of financial markets. From predicting market volatility to pricing intricate financial derivatives, these models offer quantitative analysts and traders the tools to understand, measure, and act on the diverse risk and opportunity landscapes. This document delves into three pivotal financial models: the Heston Model, Jump-Diffusion Model, and the GARCH Model. Introduced by esteemed financial mathematicians and practitioners, these models stand at the forefront of quantitative finance, bridging theory with practical application. With the evolution of technology, especially the advent of quantum computing, their relevance has only augmented. Let's embark on an explorative journey into the depths of these models, uncovering their mathematical intricacies, applications, and future implications.
Quantative Financial Modeling.pdfPage updated
Google Sites
Report abuse