Context

The dataset was created to predict market recession as inspired by assignment notebook in an online course, Python and Machine Learning for Asset Management by Edhec Business School, on Coursera. However, I aimed at doing this exercise for Indian economy but due to lack of monthly data for most indicators, I used FRED database similarly used in the course.

Content

The time period chosen is 1996-2020 according to most data available.

Acknowledgements

1. Python and Machine Learning for Asset Management by EDHEC Business School, www.coursera.org for recession period data formed according to NBER. 2.FRED, fred.stloiusfed.org for macro-economic indicator data
2. Business cycle dating, www.nber.org
3. Yahoo Finance for stock market data
4. Macrotrends.net for PE ratio
5. quandl.com for snp dividend yield

Inspiration

This dataset is inspired by the assignment notebook in the online course mentioned to predict market recession for portfolio management.

The provided Python code is developed to extract data from the Federal Reserve Economic Data (FRED) regarding Bachelor's or Higher degree education in the United States, specifically at the state and county levels. The code generates data based on the current date and is available up until the year 2021.

This code is useful for research purposes, particularly for conducting comparative analyses involving educational and economic indicators. There are two distinct CSV files associated with this code. One file contains information on the percentage of Bachelor's or Higher degree holders among residents of all USA states, while the other file provides data on states, counties, and municipalities throughout the entire USA.

The extraction process involves applying different criteria, including content filtering (such as title, frequency, seasonal adjustment, and unit) and collaborative filtering based on item similarity. For the first CSV file, the algorithm extracts data for each state in the USA and assigns corresponding state names to the respective FRED codes using a loop. Similarly, for the second CSV file, data is extracted based on a given query, encompassing USA states, counties, and municipalities.

Dataset Overview:

This dataset offers a comprehensive time series analysis of three vital economic indicators in the United States: Gross Domestic Product (GDP), Unemployment Rate, and Consumer Price Index (CPI). Spanning from January 1974 to January 2024, this dataset provides valuable insights into the U.S. economy over the past five decades, capturing periods of growth, recession, and inflation.

Contents:

• GDP Data (gdp_data.csv): Quarterly data on the Gross Domestic Product, measured in billions of dollars, highlighting economic performance and trends over the years.
• Unemployment Data (unemployment_data.csv): Monthly data on the unemployment rate, showing fluctuations in labor market conditions and workforce participation over time.
• CPI Data (cpi_data.csv): Monthly data on the Consumer Price Index for All Urban Consumers (CPI-U), capturing changes in the price level of consumer goods and services and reflecting inflationary trends.

Usage and Applications:

• Economic History Analysis: Examine long-term trends and cycles in U.S. economic performance, including periods of recession and expansion.
• Predictive Modeling: Develop models to forecast future economic conditions based on historical data patterns.
• Policy Impact Studies: Analyze the effects of fiscal and monetary policies on GDP, unemployment, and inflation over time.

Data Sources:

The dataset is sourced from the Federal Reserve Economic Data (FRED) database, maintained by the Federal Reserve Bank of St. Louis. FRED is a comprehensive resource for economic data, widely used by researchers, analysts, and policymakers.

How to Use the Dataset:

• Exploration: Utilize tools like Pandas and Matplotlib in Python to explore and visualize the dataset.
• Time Series Analysis: Apply techniques such as ARIMA, exponential smoothing, and seasonal decomposition to analyze trends and seasonality.
• Comparative Studies: Compare economic performance across different decades and investigate interactions between GDP, unemployment, and CPI.

Note: This dataset is intended for educational and research purposes. Users are encouraged to cite the original data source (FRED) when using this dataset in publications or presentations.

This dataset was download via the Python library FREDAPI.

This data is part of my project with KaggleX. I hope this dataset may be of use to you.