Skip to content

DeveloperGuide.md

Latest commit

 

History

History
27 lines (20 loc) · 1.37 KB

DeveloperGuide.md

File metadata and controls

27 lines (20 loc) · 1.37 KB
Raw

Recession Predictor (0.1.0)

For Developers

Code Walkthrough

The following text walks through the main functionalities of the modules.
For instructions on how to run the code, click here.

make_dataset.py

Gets raw data from the FRED API and Yahoo Finance. The code to get data from Yahoo Finance comes from this StackOverflow post. This module also creates most of the features to be used later on in the analysis.

build_features_and_labels.py

Builds some additional features, and organizes all the raw data into the final dataset to be used in the rest of the analysis.

exploratory_analysis.py

Creates charts for exploratory analysis, and saves all charts into the exploratory.pdf file.

testing.py

Runs backtests for each model. Model-specific code is stored in the /models/ folder.

test_results.py

Plots model predictions (probabilities) in line charts, for all models.

deployment.py

Runs the chosen model in real-time. Model-specific code is stored in the /models/ folder.

deployment_results.py

Plots model predictions (probabilities) in line charts, for the chosen model. Also outputs a deployment_chart.csv file containing the chosen model predictions.