EXP-010 design — Connors RSI(2) daily, generalization across symbols
- Work week:
2026-W36 - Run date:
2026-09-04 - Status: run completed ·
hold - Baseline: the Connors RSI(2) daily contract of
EXP-005/007
Purpose
Check whether the RSI(2) result observed on the single ETF A069500 reproduces on other symbols. The symbols are fixed to the data-quality-passed core basket of EXP-008 before seeing any RSI results. This experiment covers daily bars only and is not mixed with the 5-minute EXP-006/009 results.
Scope
| Group | Symbols |
|---|---|
| Control | A069500 |
| Generalization targets | A000660, A012330, A035420, A051910, A055550, A068270 |
Fixed contract
- Signal: enter at the next trading day's open when the close is above SMA(200) and RSI(2) < 5
- Exit: exit at the next trading day's open when RSI(2) > 70
- Comparison set: baseline, -7% hard stop, 5-bar time exit, both conditions combined
- Capital: initial 3,000,000 KRW per symbol and variant, 95% of cash just before entry, integer quantities, sequential compounding
- Risk: no blocking for the strategy-edge diagnosis; for the account-behaviour diagnosis, daily -3%, monthly -8%, -20% from peak
- Periods: train from each dataset's start to 2023-12-31, validation 2024, locked OOS 2025-01-01~2026-08-26
Cost profiles
| Profile | Commission | Slippage | Tax |
|---|---|---|---|
zero | 0bp/one-way | 0bp/one-way | None |
actual_proxy | 1.5bp/one-way | 0bp/one-way | Stocks per period schedule, ETFs none |
actual_proxy_slip2 | 1.5bp/one-way | 2bp/one-way | Stocks per period schedule, ETFs none |
actual_proxy is a provisional value before checking actual CREON fills, and 2bp one-way is a sensitivity stress.
Decision criteria
- Check whether positive results repeat on other symbols at baseline costs and under the 2bp stress.
- Do not adopt on the count of positive symbols alone; look at trade counts, OOS CAGR, MDD and year/trade concentration together.
- If OOS trades are fewer than 30 per symbol, keep
holdfor lack of generalization evidence. account_haltresults are a behaviour diagnosis of the account stop policy and are not used as the main result for strategy edge.
Outputs
- Run:
scripts/backtest/run_connors_rsi2_daily_basket_exp010.py - Results:
reports/experiments/EXP-010-connors-rsi2-daily-basket.md - Charts:
reports/experiments/EXP-010-connors-rsi2-daily-basket.html - Data manifest:
reports/experiments/EXP-010-data-manifest.json - Summary ledger:
reports/ledgers/EXP-010-summary.csv - Trade and signal ledgers:
reports/ledgers/EXP-010-daily-*-*.csv
Source and migration evidence
Source: docs/test-plans/2026-W36-connors-rsi2-daily-basket-exp010.md
Original SHA-256: e5f056d0eaa3395b090cbdf9790b0f495bc3f25d1a2a540e43c5b48457e589e6
Raw data and account information are not part of the public package. The original results do not replace statistical or scientific revalidation. The English text is a translation; numbers, tables and charts are carried over unchanged.