Niall Murray

NIALL MURRAY CAREER BACKGROUND BLOOMBERG INTELLIGENCE Msc Econometrics PROFILE Ambitious, Econometrics student who possess a strong analytical skill set with a willingness to learn and develop. Enjoys operating within and across teams and believes that any goal can be achieved with effective communication and a strong work ethic. Infosys | April 2017-July 2018 • Liaise with Bloomberg Analysts in EMEA and APAC across multiple sectors, primarily focused on Oil & Gas as well as Financials. • Construct 3 statement financial models; advanced financial statement analysis and identifying key performance metrics for each company. • Work with fundamentals & macro-economic statistics to ensure data comprehensiveness and quality control for existing and new data of target companies. • Engage with data and other global teams to develop and enhance content and functionality across the terminal. REINSURANCE ANALYST Ironshore | November 2016-April 2017 SKILLS Python, R, Excel, SQL Machine Learning Econometrics Data Cleaning & Manipulation Pandas, Numpy Keras, Sklearn Matplotlib, Requests Teamwork Attitude Problem Solving • Review a range of reports produced by outsourced data centres in India and the US to ensure data has been compiled correctly and identify any possible discrepancies or irregularities. • Analyse monthly ceded reinsurance results to identify payment patterns and any anomalies. • Assisting with the payment of reinsurance premiums in accordance with the terms of the contracts. EDUCATION & TRAINING TILBURG UNIVERSITY Master of Science, Econometrics and Mathematical Economics -) CONTACT INFORMATION Telephone: - Email:- Notable Modules: Panel Data Analysis and Microeconomic Decisions: 65% Empirical Finance: 80%, Econometrics 3: 60% Data Science: 65%, Econometrics 1: 65% NATIONAL UNIVERSITY OF IRELAND GALWAY Bachelor of Science, Financial Maths and Economics -) Notable Modules: Economics of Financial Markets Seminar 2: 92% Actuarial Mathematics: 75%, Monetary Economics: 67% Discrete Mathematics: 67%, Non-Linear Systems: 77% TECHNICAL EXPERIENCE Examining the Profitability of Technical Analysis in the Broader Cryptocurrency Market (2019) Investigated the performance of over 10,000 trading rules on the two different time horizons in the cryptocurrency market. Utilized Whites Reality check to determine whether profitable rules are statistically significant. Also using these trading rules for comparison, I ascertained the efficiency of the cryptocurrency market vs traditional markets. Machine Learning Trading Model (2019) Working with cryptocurrency data I have used both LTSM and XGboost models to build low and high timeframe models which go either long or short. Over the past year I have iteratively improved my models through adding a variety of new features and changing my objective function to penalize losses more. Initially my models began to classify my label at an accuracy range 50%-60% which is now in the range of 60%-70%. To bring these models into production I use AWS to automate their trading. Optimal Model to Forecast Football Players Wages (2019) Using R and retrieving my data from Kaggle I applied a variety of models such as linear regression, lasso, ridge and random forest to find which model had the best performance in terms of OOS MSE for forecasting a football player’s wages. The hyperparemeters were tuned using k fold cross validation, the lasso model was found to outperform the others likely due to the other models suffering from too many irrelevant features which may overfit the models to the training sample. An Analysis of Stock Returns Across Industries (2018) Operating in a group of three we retrieved daily, monthly and yearly returns for the past 100 years on portfolios made up of stocks of for multiple industries. Using this data, we verified the existence of certain stylized facts in the market such as volatility clustering and also tested whether the market can be predicted using calendar effects. We then tested the effectiveness of CAPM and Fama-French 5 factor model at explaining the momentum factor. Time Series Analysis and Forecasting of Oil Prices (2016) Built a model to forecast price movements of WTI crude oil in the short and medium term. Collecting interrelated datasets such as the number of oil rigs operating in the US, interest rates and the price of oil futures over the same 15-year timeline I obtained clearer understanding of what indicators were influencing the price of crude within that timeframe. Through the coding language R, I used various statistical techniques such as OLS and ARIMA to infer relationships and predict the future price.