Close Menu
    [OC] A real-time dashboard tracking the 6 major US Recession Indicators (Yields, Unemployment, Production). Current risk is 21%.
    Data

    [OC] A real-time dashboard tracking the 6 major US Recession Indicators (Yields, Unemployment, Production). Current risk is 21%.


    Data Source:

    • FRED (Federal Reserve Economic Data): Series SAHMREALTIME (Sahm Rule), DGS10/DGS2 (Treasury Yields), ICSA (Jobless Claims), INDPRO (Industrial Production), HOUST (Housing Starts).
    • Univ. of Michigan: Consumer Sentiment data via FRED (UMCSENT).

    Tools Used:

    • Frontend: React, TailwindCSS, and Recharts for the visualization.
    • Backend: DataSetIQ API (my own project) to normalize the different reporting frequencies (daily vs. monthly) into a single live score.

    Methodology: I created a composite index (0-100) where 0 is a booming economy and 100 is a guaranteed recession. The model weights historical leading indicators:

    • 50% Weight: Yield Curve + Sahm Rule (Highest predictive power).
    • 30% Weight: Jobless Claims + Housing Starts.
    • 20% Weight: Industrial Production + Consumer Sentiment.

    Key Insight: The current score is 21% (Low Risk). There is a massive divergence right now: Consumer Sentiment is flashing "Recession" (-25% YoY), but hard data like Industrial Production (+1.5%) and the Yield Curve (Positive +0.60%) suggest a soft landing is holding.

    Live Interactive Dashboard: You can check the live data and the historical backtest here: Recession Risk Index

    Also, please do let me know if there are areas to improvise here.

    Posted by dsptl

    View 9 Comments

    9 Comments

    1. Jpbbeck99 on

      The thing you’re not taking into account here is the Median Real Wages vs. Productivity, you have productivity accounted for but production can be done without stimulating the economy if you don’t pay people to do it. Productivitiy has skyrocketed since Covid due to automation and computing while median real wages have stagnated and even dropped in the same time period. The other economic indicator you should be looking at is the Gini Coefficient. The U.S. the second highest first world country after turkey and the percents for the gini coefficient have been steadily increasing since the 60’s with a rapid increase in the past 20 years across the world.

      TLDR: your model fails to display the reverse pyramid economy because you aren’t accounting for cost of living vs real wages for the majority of the western world.

    2. lightning228 on

      Risk: 8.60000000000000003. Nice.

      In all seriousness, just add a tofixed(2) or something similar on that

    3. blueskiess on

      It’s nice and I’m familiar with it as I work in finance. You’re asking someone to pay for an index when the underlying data is free?

    4. Gibodean on

      Is “having a lunatic grifter as president” one of the indicators? Because I think it should could for at least 2 of the other ones.

    6. darkexistential on

      can you create data without the top earners included… i forget where i saw it but i remember reading an article from the fed or something that it’s the rich that’s carrying the economy and their spending that allows growth while the rest are just living in a recession

    7. AnthraxCat on

      Your Best Case | Expected | Worst Case blocks are inverted, unless you think the best case scenario is a sooner recession?

      I am also curious, since you indicated some of the 60 years of backtesting, did you also do any checks for non-recession years?

    9. dragnabbit on

      A stable housing market is almost certainly not going to happen. Tariffs on certain building supplies, plus 50% of the home-building workforce deported… It’s hardly the makings for a steady industry in the short term.