Business Analytics

Purpose
The aim of this document is to help you decide your term project topic, form your research question(s), and check out the related databases links to confirm the existing data and its relevance to your selected research topic and question(s). Identifying a research question pertaining to the data sources provided may be the easiest, however, it could be the other way around. You may have a very specific research question, for instance factors influencing inflation. You can determine these factors from the existing literature and the media and then check the data sources below whether relevant data exists.

Sources for datasets
Please select data from the sites and links provided below which have been validated and are trusted sources of data/information. This information is also available on Canvas under Library Guides.

For questions regarding the websites or you need further assistance searching within a site, contact our LCOB Librarian Regina Beard (rmbeard@fgcu.edu). Please include a reference to this class in the email subject line and details of what you are looking for.

Sample size
As a rule of thumb, at any given time, 100 records or data points are needed. The exceptions to this would be the unavailability of data for some of your variables or indicators. In that case, the minimum sample size can range between and 30 and 50 data points.

The frequency of data
Whether it is collected annually, monthly, quarterly, or weekly needs to be considered to determine whether you meet the sample size rule.

So, generally,

• If a statistic is collected monthly, you need at least 8-10 years’ worth of data points resulting in 80 to 100 data points
• If a statistic is collected quarterly, you need at least 20-25 years’ worth of data points resulting in 80 to 100 data points
• If a statistic is collected annually, you need 80-100 years’ worth of data points. Specific to yearly data you may study 30-50 years of data as 80 or 100 years may not be available, also not so relevant to today.

Collection rule

Above is true for all your variables. For instance, if you want to determine the relationship between Sales and Labor force, for each of these variables you will need the same amount of data points with the same frequency and within the same time frame. For example, you need data points from 1980 to 2022 collected quarterly for both sales and labor force. This is true whether from the same database or a from two different sources. (ex. FRED and EIA are different databases).

Further
The data source will indicate whether data is seasonally adjusted or not, or given in today’s dollars. Make sure you have this information marked on your file.  Use always the seasonally adjusted and indexed data. This is especially important when you use data over a big span of time, such as 1920-2022. Also, in most cases, data may need to be studied in chunks, divided into smaller time frames: last five years, 10 years, or separated based on certain milestones and crises.

Federal Reserve Bank of St. Louis (FRED)
Economic Indicators for US and per States

• • • Gross Domestic Product (GDP) (1947-2022)
    • Gross National Product (GNP) (1947-2022)
    • Imports of Goods and Services (1947—2022)
    • Exports of Goods and Services (1947-2022)
    • Consumer Price Index (CPI)for all urban consumers, (1947-2022)
    • Producer Price Index(PPI) for all commodities (1913-2022)
    • Industrial Production (1972-2022)
    • US Labor Productivity, Manufacturing (1987-2022). See also Bureau of Labor Statistics. 
    • Inflation, Consumer prices for the U.S. (1960-2021)
    • U.S. unemployment rates, seasonally adjusted (1948-2021)
    • Real Median Income – Click Geographies to limit by state
    • Real disposable personal income (per capita, 1959-2021)

Florida Department of Economic Opportunity (state employment statistics)

Under Current Employment Statistics (CES) you will find Historical Nonagricultural Employment by Industry (1990-2021). Look for the dropdown menu in the center of page. Earnings (1998-2021).

Federal Reserve Bank of St. Louis (FRED)

U.S. Real estate

• • • Median sales of homes sold by financing type (1988-2022, quarterly)
    • Home ownership rates, (1980-2021, quarterly)
    • Houses sold by region*, (1963-2020, regions not described)
    • Median and Average Sale Price of Houses Sold (1963-2021; quarterly, monthly)
    • New Residential Construction, 1959-2021 (housing starts, by units, privately owned, more)
    • Home Ownership Rate for Florida (not seasonally adjusted)

*Check Census-defined regions to see how the Census defines a region.

This short tutorial demonstrates how to find stock prices and financial statements in  Mergent Online. Copy and paste it into your browser: https://www.screencast.com/t/3yINPF4gDk

U.S. Energy Information Agency (EIA) – energy sources, energy consumption, energy prices, and energy production. 

• • • Monthly Energy Review—a cumulative report that collects data by sector, dating back as far as 1973.  Data categories include consumption by sector, coal, crude oil, electricity, renewable, more
    • Gasoline and Diesel Fuel Update -prices by region and state; frequency, weekly and monthly, dating back as far as 1993.
    • U.S. Energy- State Energy Data System (consumption, production) by source, 1960-2019

Bureau of Transportation Statistics

Sales and production of hybrid-electric, plug-in hybrid and electric vehicle manufacture

• • • Hybrid-Electric, Plug-In Hybrid-Electric Vehicle sales (2000-2021, length of time varies)*
    • Alternative fuel vehicle and hybrid electric vehicle models, by manufacturer (1991-2019)

National Center for Education Statistics (NCES)

• Degrees conferred by level of degree and sex of student (National Center for Education Statistics), 1869-2030

Wages & Salaries per full-time equivalent employee (FTE) by industry. This information is spread over four tables under Section 6 (Income and Employment by Industry)  6.6A (1929-1948); 6.6B (1948-1987); 6.6C (1987-2000); and 6.6D (1998-2020)