Forecasting-VAR-Granger-Causality

This project focuses on the Data Science, Statistics, as well as some financial theory revolving around how financial markets behalf.

Goal

Our goals involve the following:

• Part 1: Analyze the descriptive statistics and the trends of all of our variables, which involves testing for skewness, asymmetry, and kurtosis.
• Part 2: Testing for autocorrelation which involves a process of using an autoregressive model.
• Part 3: Testing for stationarity using the Augmented Dickey-Fuller test.
• Part 4: Testing for causality with each individual variable using Granger Causality.
• Part 5: Creating the model based on the relationship with the variables.

This analysis will allow us to assess the given relationship between specific financial markets and effectively forecast them.

Data

We used data from Quandl, which allows us to extract financial and economic data for analysis in Python. We also used some data from the Federal Reserve Economic Database or FRED.

Enviroment and Tools

1. Jupyter NoteBook
2. Numpy
3. Pandas
4. Matplotlib
5. Scipy
6. Statsmodels
7. Quandl