Strategies

Style-Matched-Benchmark Alpha (GTJA 'CSI500 Alpha Touchstone', restated for US utilities): long the 3 of 10 large-cap utilities with the lowest residual volatility and lowest abnormal turnover, short XLU beta-matched, monthly rebalance (USEQ 1-DAY, long-short, market/sector-neutral)

Outcome: Abandoned

UtilityStyleMatchedAlpha

Outcome Summary

UtilityStyleMatchedAlpha tried to capture idiosyncratic-volatility and abnormal-turnover alpha in 10 US utilities, hedged with the style-matched XLU benchmark as the GTJA report suggests. Across 3 iterations it never got past verification: Layer 1.5 flagged a mismatch because the optimization plan fixes leverage at 1.6. Iteration 3 resubmitted code identical to iteration 2 without addressing that feedback, so the pipeline abandoned it at the verification loop. The backtest, optimization, analyst and risk stages were not reached, so the hypothesis itself remains untested.

Strategy report

Hypothesis

Classification: (b)/(a) borderline. The source (Guotai Junan quant report #60, 2015-06-04, via BigQuant) is an abstract of a style-neutral multi-factor A-share selection strategy hedged with CSI500 index futures. Its one concrete, transferable claim is about the HEDGE: alpha is only clean when the hedge benchmark matches the universe's industry and size distribution, so hedging a mid-cap selection with a mid-cap index ('the touchstone') removes style beta that a mismatched index (CSI300) leaves in. The abstract names no specific factors, and A-share venues (CNEQ/HKEQ) have no working data estate here (lesson L103), so we restate it on USEQ. Universe: 10 long-history S&P 500 utilities (NEE, DUK, SO, D, AEP, EXC, SRE, XEL, ED, PEG). Hedge: XLU, the benchmark built from the same industry and size bucket, so industry, rate-sensitivity and size style net out by construction. This is the paper's 'touchstone' idea. Alpha factors (OHLCV only, because no fundamentals are in the catalog): the two price/volume factors that dominate A-share multi-factor models and have US evidence too. (1) Low residual volatility: std over 60 sessions of the daily residual vs XLU (the idiosyncratic-volatility anomaly, Ang et al. 2006). (2) Low abnormal turnover: 20-session mean volume divided by 250-session mean volume (high attention or turnover predicts underperformance). Composite = average cross-sectional rank. Every 21 sessions, hold the 3 best-ranked names equal-weight long and short XLU sized to the basket's 120-session beta, so the book is dollar- and beta-neutral within the sector. Cost: zero commission, ~0.02-0.05% spread+impact per leg on mega-liquid utilities. The universe is capped at 10 names (lesson L173) so per-rebalance ranking is O(10) on precomputed rolling stats. Credit: Guotai Junan Securities quant team, 'Quantitative Topic #60: CSI500 as the Alpha Touchstone', 2015-06-04, republished by BigQuant (宽邦).

Venue choice: USEQ (5%) instead of BINANCE (63%, far over its <=40% quota). The book is long-short (15.7% vs the 45%+ we want) and multi-instrument. The mechanism is new to the corpus: the sector-matched hedge plus low-idio-vol/low-turnover selection. It is not the dead short-horizon reversal family (L113: 21-session hold, no reversal factor), not industry momentum (died 2026-10-01: it traded sector spreads vs SPY, this trades single names vs their own sector ETF), not a calendar window (L179: rebalance is a fixed cadence, but the trigger is cross-sectional rank) and not a cointegration pair (no spread z-score). Data check on 2026-10-02: SPY/XLU/XLK USEQ 1-DAY are in the catalog from 1993/1998 with ~6985 rows, and the IndustryFactor run on 2026-10-01 completed 25 years on the SPDRs, so the earlier XLK gap is resolved. All 10 utilities trade continuously since before 2001 (D, EXC, SRE, PEG, ED listed throughout).

verification_loop: code_identical: iteration 3 resubmitted byte-identical code to iteration 2 — the previous feedback was not acted on, so the result cannot change. Either change the code to address it, or state recommendation='abandon' with the reason if no code change can help. Previous feedback: Verification failed (Layer 1.5 — hypothesis/config consistency) [class=hypothesis_mismatch]: - optimization_plan fixes leverage at 1.6

Implementation

Ranks ten US utilities by residual volatility against XLU and abnormal volume, holds the three best names, and shorts XLU according to the basket’s beta on a 21-session cadence.

The Layer 1.5 feedback identifies a config mismatch: the research plan fixes leverage at 1.6. The config now uses 1.6; the previously passing strategy code is unchanged.

Verification Results

code_identical: iteration 3 resubmitted byte-identical code to iteration 2 — the previous feedback was not acted on, so the result cannot change. Either change the code to address it, or state recommendation='abandon' with the reason if no code change can help. Previous feedback: Verification failed (Layer 1.5 — hypothesis/config consistency) [class=hypothesis_mismatch]: - optimization_plan fixes leverage at 1.6

Iteration History

Verification failed (Layer 1.5 — hypothesis/config consistency) [class=hypothesis_mismatch]: - optimization_plan fixes leverage at 1.6

Outcome Summary

If verification reports a hypothesis/config mismatch such as fixed leverage, either change the strategy config to match the optimization plan or recommend abandoning; resubmitting the same code always ends the run.

Verification Layer 1.5 (hypothesis/config consistency, class=hypothesis_mismatch) failed with 'optimization_plan fixes leverage at 1.6'. On iteration 3 the developer resubmitted code byte-identical to iteration 2, so the unchanged-code check ended the run as a verification_loop abandonment.

A US-utilities restatement of GTJA's 'CSI500 Alpha Touchstone' idea: each 21 sessions, go long the 3 of 10 large-cap utilities with the best combined rank on low 60-session residual volatility vs XLU and low abnormal turnover, and short XLU sized to the basket's 120-session beta so the book is sector- and beta-neutral (USEQ, daily bars).

There are no performance figures. The strategy never passed verification, so the backtest report is empty and no optimization, Sharpe, return or trade counts exist.

Analysis

数量化专题之六十：中证500之阿尔法验金石-国泰君安-20150604 - BigQuant量化交易

Backtest and paper results are hypothetical. Trading involves risk of loss.