What it does
The code turns tactics into prices and can substitute coded probabilities when live calls are unavailable. Reported results describe a model-plus-policy system, not financial advice.
Benchmarks & research
Continuous Double Auction AI Market Benchmark
Compare simulated traders that sample tactics from Jev probabilities with random traders, using a dashboard and saved market results.
Bookmark: Continuous Double Auction AI Market Benchmark
Keep this in your collection.
Leave a noteWhat would you try with this?
: Continuous Double Auction AI Market Benchmark
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
Begin with the saved notebook and results to study an auction where buyers and sellers submit prices. Jev assigns chances to four trading tactics; code uses those chances to pick one and calculate a price. This can teach a simulation design, but it is not a tool for choosing real investments.
Ask a developer to restrict the dashboard to your computer before opening it; its default network settings can expose controls without sign-in. A developer needs Python tools and a TypeSafe access key for live calls. Other runs can use coded rules instead. Check which produced the displayed decisions. The spending setting tells later work to stop; it cannot guarantee a total bill.