AI for economics research is only as good as the data it can show you the source for. AI for Research looks up public economic data from FRED, the OECD, the World Bank and US Treasury Fiscal Data, and company filings from SEC EDGAR, all free in Official Sources. The Financial research workbook keeps each reported metric with its period, unit, currency and the passage it came from. Data Lab analyzes your series in the browser, and Deep Research finds and cites the literature.
Sources for business and economics
| Source | What it holds | Where in the product | Credits |
|---|---|---|---|
| FRED | US and international economic time series: output, prices, rates, employment | Official Sources | 0 |
| US Treasury Fiscal Data | Federal debt, receipts, spending and interest rates | Official Sources | 0 |
| OECD | Economic and social statistics for member countries | Official Sources | 0 |
| World Bank | Development indicators by country and year | Official Sources | 0 |
| SEC EDGAR | US company filings, such as annual and quarterly reports | Official Sources | 0 |
| OpenAlex and Semantic Scholar | 250M+ scholarly works, including economics and management journals | All papers; Find Papers | Search 0 |
| Web | Central bank reports, working papers, statistics office releases | Web source in Deep Research | As the run's depth |
Deep Research costs by depth: Summary 1 credit, Deep 15, Detailed 45 (paid plans).
Tools for business and economics
| Tool | What it does | Credits |
|---|---|---|
| Official Sources | Look up the series and filings above, with their source and units | 0 |
| Financial research workbook | Reported metrics with company, definition, value, unit, currency, period, filing date and the source passage; scenario assumptions kept separate; a research memo | 0 |
| Data Lab | Correlation, regression, group comparisons, calculated in the browser; PDF report | 0; AI write-up 2 |
| Bayesian statistics | Bayesian t-tests and proportions with Bayes factors | 0 |
| Deep Research | Cited answers from the economics literature and the web | Summary 1, Deep 15, Detailed 45 (paid) |
| Find Papers | Search journals and working papers | 0 |
| White paper writer | An industry or policy white paper built on cited evidence and your own figures | Whole 15, section 3 |
The Financial research workbook is in Workbooks. Its comparison view only groups metrics that share the same definition, period, unit, currency and version (as reported or restated), so a quarterly figure is never compared with an annual one by accident.
A worked example
Example (invented for illustration): an economics undergraduate is writing a dissertation on whether rising mortgage rates slowed US housing starts.
- In Official Sources he looks up FRED's 30-year mortgage rate and housing starts series, monthly, and downloads both. 0 credits.
- He adds OECD house price data for comparison with three other countries. 0 credits.
- A Deep Research Summary answer on "how do mortgage rates affect housing construction?" cites eight papers. 1 credit. A follow-up on lags: 1 credit.
- In Data Lab he runs a Pearson correlation and a linear regression of starts on the rate lagged six months, which he computed in his spreadsheet. Free. The AI results paragraph costs 2 credits, and he downloads the PDF report, free.
- For one homebuilder case study, he records revenue and orders from two SEC EDGAR annual reports in the Financial research workbook, each with period, unit, currency and the quoted passage. 0 credits.
Total: 4 credits, inside the Free plan's 45 a month. His supervisor asks for a stronger literature section, so he runs one Deep report: 15 more, 19 in all.
Citation styles and reporting standards
Economics journals mostly use author-year styles, such as the AEA's or Chicago author-date, and business schools often ask for APA or Harvard; all are among the 10,000+ styles in the editor. Data needs a citation too: give the series name and identifier, the publisher, the frequency, and the date you downloaded it. For company data, cite the filing, its form type and the period. The SAMPL guidelines in the guideline check help with reporting regression results.
What it does not do
- It is not investment, tax or accounting advice, and it does not recommend securities.
- No paid market data: Bloomberg, Refinitiv, Statista, Compustat and CRSP are not searched, and neither are paid market research reports.
- No econometric models beyond regression: no panel data, IV, ARIMA or VAR.
- It does not forecast. Projections appear only when a cited source makes them, labeled as that source's.
- Data series get revised. Note the vintage or the date you downloaded each series.
Who uses it
- Undergraduates and master's students: see AI for undergraduates and AI for master's students.
- Economists and business school faculty: see AI for academics.
- Analysts and strategy teams: see AI for industry.
For surveys, interviews and social data, see social sciences.
Learn more
- FRED economic data for research
- How to write a white paper
- Which statistical test to use
- How to find sources for a research paper
Last updated 2026-11-13