Analysis tool for tracking economics job market postings from the American Economic Association's Job Openings for Economists (JOE).

• Cumulative posting trends by week across multiple years (2015-2025)
• Rolling 4-week flow analysis to smooth out weekly fluctuations
• Multiple filter views:
  • Overall market
  • Finance jobs (based on JEL G classifications)
  • Federal Reserve and bank regulator positions
  • US vs. Non-US jobs
  • Regional analysis (US, Canada+Europe, Asia)
  • Job type (Tenure Track, Non-Tenure Academic, Industry)
• Interactive HTML plots and static PNG charts
• Current week interpolation for more accurate mid-week comparisons

pip install pandas matplotlib plotly seaborn openpyxl

Run the standard analysis without interpolation:

python3 analyze_jobs.py

This generates:

• Static PNG charts for all categories
• Interactive HTML visualizations
• Summary statistics for each year

Run analysis with current week interpolation:

python3 analyze_jobs.py --interpolate

When the --interpolate flag is used, the script:

1. Detects the current day of the week (Monday = day 1, Sunday = day 7)
2. Calculates what fraction of the current week has elapsed
3. Projects the current week's posting count to a full-week estimate
4. Updates all 2025 cumulative totals accordingly

Example:

• Current day: Monday (day 1 of 7, or 14.3% of week)
• Postings so far this week: 15
• Interpolated full-week count: 15 / 0.143 ≈ 105
• Added to cumulative: +90 postings

When to use interpolation:

• Mid-week comparisons to understand if 2025 is tracking ahead/behind prior years
• Blog posts or reports generated during the week
• Real-time monitoring of market trends

When NOT to use interpolation:

• Final end-of-season analysis (use actual data)
• Early in the week (Monday/Tuesday) when estimates are less reliable
• When postings are clustered at specific times during the week

Note: The interpolation assumes an even distribution of postings throughout the week, which may not reflect actual posting patterns. Use with appropriate context.

All outputs are saved to the current directory:

Without interpolation:

• job_postings_by_week.html - Interactive cumulative chart
• job_postings_rolling_4wk.html - Interactive rolling 4-week chart
• Plus category-specific files (finance, fed, us, non_us, regions, job types)

With interpolation:

• All files get _interpolated suffix (e.g., job_postings_by_week_interpolated.html)

The script expects Excel files in a data/ directory with the following columns:

• Date_Active - When the posting became active
• JEL_Classifications - JEL classification codes
• jp_institution - Institution name
• jp_section - Job posting section (Tenure Track, Nonacademic, etc.)
• locations - Geographic location information

All postings across all categories

Filtered to positions with "G - Financial Economics" JEL classification

Positions at:

• Federal Reserve Banks (all districts)
• Board of Governors
• FDIC
• OCC

• US vs. Non-US: Split between United States and international positions
• Regional: US, Canada+Europe, and Asia breakdowns

• Tenure Track: Tenure-track or tenured positions
• Non-Tenure Academic: Visiting, temporary, part-time, adjunct positions
• Industry: All nonacademic positions

• Academic year calculation: Starts from week 31 (early August)
• Week numbering: ISO week numbers, adjusted relative to start of academic job market
• Interpolation: Linear projection based on elapsed days in current week
• Rolling window: 4-week moving average for smoothing

Created for tracking the economics job market. Data sourced from the AEA's Job Openings for Economists.