An Intro to Myself...
Please do get in touch!
- - - - - - - - - - Project 1 - - - - - - - - - -
Backtesting Machine learning Trading Strategies
Manipulating time series data, creating, testing and tuning complex machine learning tradings algorithms with multiple hyper paramter spaces
Downloading and constructing time series data, back testing multiple classification machine learning algorithmns as trading strategies. Analysing the algorithmns risk and return characteristics including maximum drawdown, drawdown period and VAR .
- - - - - - Project 2 (Part B) - - - - - -
creating a derivatives package via object orientated programming
Using the power of python to create a scalable, dynamic derivatives package for valuing european options
Leveraging the power of inheritance and OOP to create a dynamic financial modelling package with a structured financial environment to track and record parameters whilst having the capability to accept various user built valuation and projection algorithms classes
- - - - - - Project 2 (Part A) - - - - - -
Valuing derivatives with Monte Carlo simulation of Stochastic Processes
Simulating multiple stochastic processes including Black-Scholes Merton and Heston's Stochastic Volatility to value options
Understanding random number generation, stochastic process projection, cholesky correlation and least squares monte carlo simulation to create functions for valuing both european and american options
- - - - - - - - - - Project 3 - - - - - - - - - -
Modern Portfolio Theory and utilising the python scipy package
Creating the efficent frontier and capital market line for a portfolio of financial instruments utilising scipy
Constructing portfolios using modern portfolio theory, plotting the efficient frontier and solving for the maximum sharpe portfolio using scipy to formulate the efficient frontier function and capital market line
- - - - - - - - - - Project 4 - - - - - - - - - -
Tree based machine learning models and ensembles
Creating Decision Tree and ensemble ML workflows for regression and classification problems and investigating bagging and boosting
identifying the most predictive parameters within fitted models, carrying out gridseaches, cross validations and hypertuning on an array of different classifiers and understanding performance metric enhancement
- - - - - - - - - - Project 5 - - - - - - - - - -
Unsupervised Learning - Clustering and Dimension Reduction
Applying Kmeans clustering and dimension reduction by principal component analysis to unlabelled data
Utilsing clustering to make machine learning predictions and understanding dimension reduction to find the intrinsic dimension of a data set
- - - - - - - - - - Project 6 - - - - - - - - - -
Times Series and Stock Indices Analysis and Visualation
Manipulating time series data, investigating financial market correlations and undertaking technical analysis techniques
Utilising plotly in jupyter to create interactive charts, visualing time series correlations, creating bollinger bands and moving averages
- - - - - - - - - - Project 7 - - - - - - - - - -
supervised learning - Predicting Credit Card Approvals
Analyising the Predictability of variables in a Machine learning workflow for credit card approvals
Classification problem for predicting credit card approvals analysing the overall performance, identifying the most predictive features and hyptertuning model parameters
- - - - - - - - - - Project 8 - - - - - - - - - -
Nobel Prize Winners Characteristics Analysis
Data manipulation, analysis and visualisation
Utilising Pandas and Seaborn to investigate how the characteristics of Nobel prize winners have developed over time
- - - - - - - - - - Project 9 - - - - - - - - - -
Google Apps Competitor Landscape Analysis
Data manipulation, analysis and visualisation
Utilising Pandas, Seaborn and Jupyter Plotly functions for exploring the google apps market place to identify competitor denisty in different genres and understand pricing dynamics