Projects Page (Work in Progress):

More information about projects along with the reports are being added

• Aguared - (Master of Engineering project) at Cornell. A software engineering project in a team of 8 done in collaboration with the School of Civil and Environmental Engineering at Cornell University. The Aguared software helps engineers develop and validate water supply networks on the fly from GPS survey data. Aguared is currently being tested in the field at Honduras.  Click to download the latest version from Cornell's Gforge site here or get my copy of the installer along with the user documentation here. To just see the user documentation which contains screenshots -  click here (incase if you do not wish to install and demo the software). Technical Skills: C# programming (VS.Net-Windows Environment).

• KDDCup - Machine Learning Project -  This project involved the use of a combination of Decision Trees, Neural Networks and Support Vector Machines to make predictions on a dataset. We employed techniques such as Feature Selection, Bagging and Boosting to improve performance based on RMSE, Accuracy and ROC metrics. Click here to download report. Technical skills: MATLAB, PERL, TCSH, C programming, Linux and Windows Environment.

• Technology Rich Mutual Funds and their performance during and after the “Tech Bubble” - This paper details the results of a study of mutual funds before and after the crash of tech stocks in early 2000. The objective is to see if an investor could view the quarterly holding information provided by the different mutual funds and buy into mutual funds which had a high percentage of Technology holdings and sell mutual funds which had a low percentage of technology holdings. Click here to see the report. Technical skills: PHP, Stata, MS-Access, MS-Excel, Windows Environment.

• Pricing American and European Options using the Crank Nicolson method - In this project, I valued European Calls and Puts using the Crank-Nicolson Finite Difference method. In the case of the American Put,  the Linear Complementarity Formulation (LCF) was along with the Crank-Nicolson method using the Projected Successive Over-Relaxation (PSOR) technique. Click here to see report and accompanying code. Technical skills: MATLAB, Windows Environment.

• Organic Molecular Shuttles based on Rotaxanes -  Seminar Presented on Organic Molecular Shuttles synthesised using Self-Assembly techniques, which have potential applications in Molecular Computing as logic gates and in mimicking biological process such as ATPase. Click here to read the abstract.