For the completion of the Panel Data Econometrics module in MSc. Financial Economics. This assignment involved a recreation and extension of the seminal 2000 paper by Burnside and Dollar. Using Python, I performed data visualisations on the Burnside and Dollar dataset, implemented a variety of econometric models and discussed the links between effective development aid and growth. 

This project required a large amount of data presentation and comment on descriptive statistics, such as plotting residual distributions, box plots and variations over time. 

The repository for this project including the Jupyter notebook is available here.

Completed in May 2023 for the completion of my MSc. Economics (PSME) this thesis involved the design, calibration and simulation of an overlapping generations model for a typical developing economy. The OLG model followed the usual structure of a representative household and firm who maximise their utility and profit respectively, and I solved for three dynamic equations for the evolution of health, physical capital and education.

Calibration of the model involved the collection of household parameters such as the intra-household bargaining of men and women, the marginal propensity to consume and, among others, the relative preferences for health, education and saving. Household parameters are far more difficult to obtain than macroeconomic parameters and often involved using household surveys from developing economies or finding the parameter values in previous literature. Throughout this project, I showed my ability to work with sometimes intractable data sources to produce meaningful results.

Overall, I showed that foreign aid augments the rate of convergence of the economy to the steady state, which in turn depends on the level of gender inequality in the economy (determining the relative preferences for consumption, education, health and saving).

The GitHub repository including the MATLAB code for solving the model and calibrating the model using Dynare is available here.

This project was completed as part of the Asset Pricing module for MSc. Financial Economics. This project was in three parts, all of which were completed using Python. 

Part 1 saw the estimation of a multi-factor asset pricing model using endogenous and exogenous factors to estimate the multi-beta relationship for the S&P 500. The factors employed, in addition to the Fama-French factors, included inflationary surprise, oil price shocks, and interest rate shocks.

Part 2 involved a fundamental analysis of a stock's price using the present value model and the Gordon-Shapiro model using estimation of Vector Error Correction Models to account for the cointegrating relationship between prices and dividends. 

The final part involved Monte Carlo simulations to derive the fair price of a lookback put on the S&P500.

The GitHub repository is available here.

This project was again for the completion of MSc. Financial Economics and required use of both Python and EViews. The project involved decomposing financial series into a cyclical component, a seasonal component and a stationary component. An Autoregressive Moving Average Model was estimated for the cyclical component. This project also involved performing unit root tests, the estimation of VAR and VECM models, and identification of cointegrating relationships.

The final empirical application for this project involved the estimation of a Logistic Smooth Transition Regression model which I completed using EViews. LSTR models allow the modelling of regime changes in financial series.

The repository including the code for this project is available here. 

This project, for the completion of MSc. Financial Economics, involved the implementation of Moody's RiskCalc methodology in Python. I constructed a range of models that relied on up to 15 financial indicators in firm-level data to estimate credit default risk. I split the data into a training dataset and a test dataset to estimate the models and then assess the accuracy of the prediction.

The repository including the code for this project is available here. 

This project I completed in December 2022 for the completion of my MSc. Economics. As part of the Macroeconomics module I was required to model a country of my choice using the Real Business Cycles model. I collected data from a number of sources including GDP growth, interest rates, wages and inflation. I calibrated the model using the data and performed step-ahead forecasts to model the evolution of the variables for following quarters. This project involved working as a team and I delegated tasks according to my team's strengths. I was awarded a grade of 100% for this project.

The repository containing my code for this project is available here. 

This is the first large research project I completed. Throughout this project, I taught myself the mechanics of overlapping generations models, familiarised myself with MATLAB and Dynare, and strenghthened my skills of data collection and cleaning. This dissertation provided the bedrock of my M1 MSc. Economics thesis into foreign aid and growth and it is where I learnt the basics of MATLAB and computable general equilibrium modelling.

I simulated using my model the effect of different policy experiments by the government of Nigeria on the macroeconomy. Experiments included increasing the share of public spending to infrastructure, increasing efficiency of public spending, household transfers to reduce the cost of child rearing, a reduction in gender bias in employment, increasing mothers' household bargaining power, and equalisation in the time allocated to boys and girls during child rearing.

The repository including the model files is available here.

This project, consisting of two parts, was the final project for the Applied Econometrics course at Durham University. Using Stata, I was required to employ macroeconomic time series of my choice to try and forecast recessions (as given by the NBER recession indicator). 

The second part of the analysis involved employing econometric models to forecast the volatility and returns of three cryptoassets of my choice.

I built the core skills of data cleaning, manipulation and visualisation during this project, skills which have been used in every project I have completed since.