neilyejjey1999 / yejjeyprojects Goto Github PK

FIN5615_Project_2 is one where I utlized a simple random walk in order to model stock prices, followed by comparing the random walk model with the Black-Shoeles model.

FIN5615_Project_3 is a continuation of the previous project.

FIN5615_Project_5 is an options pricing calculator built on Python, primarily showcasing the stochastic volatility of the greeks in the model.

FIN5615_Project_6 is a robust analysis on the returns of the S&P 500, by taking a dataset of the returns from Excel using read.csv to perform a variety of performance measures using functions from the numpy library inlcuding array, argmin/argmax and argsort.

FIN5622_Project_1 utlizes machine learning to predict the price of bitcoin using a fitted linear regression model.

FIN5622_Project_2 utlizes machine learning to build a model predicting the default probability of a loan using a fitted logistic regression model.

FIN5622_Project_3 is similar to project 2, however it utilizes random forests to predict default probability as apposed to a fited logistic regression model.

FIN5622_Project_4 uses clustering in order to detect regimes in the stock market.

FIN5622_Project_5 uses TensorFlow to construct a neural network that detects the probability of a loan being defaulted on.

FIN5622_Project_6 is a further application of TensorFlow to construct a neural network which predicts stock prices based on historical data and performance.

Project 1 Questions is a series of R questions (using RStudio as the IDE) that include using R to conduct a variety of tests on imported Excel data (primarily regression, T Test, and plotting) to deterime regression equations, chi values, T value/T Test scores, and P value.

Project 2 consists of two questions, with the first examining how firms’ stock returns (y) are affected by different independent variables using multiple linear regression models. Question 2 examines if attending an after-school program has a causal effect on the improvement of students’ exam scores (y) using a difference-in-difference model.

Financial Time Series Project (1) consists of using a variety of tests in R including ACF, Unit Root Test, and fitting data into a univariate time series model to determine if Excel imported data is stationary or not.