Exchange Rate is one of the daily topics that captures everyone’s attention, especially in the current Covid-19 situation where the global economy is severely affected. Therefore, I’m highly interested in analyzing the exchange rate pattern in the past so as to be able to forecast the value in the future.
As part of this project, 3 accompanied blog posts on Medium can be found at the following links for your reference:
- Exchange Rate Prediction: EDA & Data Visualisation
- Exchange Rate Prediction: Machine Learning with 5 Regression Models
- Exchange Rate Prediction: Time Series Forecasting with ARIMA
This dataset was generated on the Federal Reserve's Download Data Program and shared publicly via Kaggle. However, further data points were added in during the analysis to tune the model for better prediction, which can be found along the codes as well as the blog posts.
As a summary, the dataset provides the exchange rates across countries which have been converted to USD for consistency and processing.
- pandas
- numpy
- matplotlib
- seaborn
- bokeh
- scipy
- sklearn
- statsmodels
- Part 1: Explanatory Data Analysis (EDA) & Data Visualisation (Bonus: Hypothesis Testing)
- Part 2: Machine Learning with 5 Regression Models
- Part 3: Machine Learning: Time Series Forecasting with ARIMA
