← All tasks
T16 · Volatility Management

Volatility-Managed Portfolios

Alan Moreira and Tyler Muir, 2017 · The Journal of Finance

64%
pass@1 across 16 systems
14/16
systems with ≥1 successful attempt
5/5
5/5 attempts passed for the best system
80 attempts · 16 systems × 5
Claude Code + Opus 5
60%
Claude Code + Opus 4.8
100%
Claude Code + Opus 4.7
100%
Codex + GPT-5.6-sol
100%
Codex + GPT-5.5
80%
OpenCode + Kimi-K3
100%
OpenCode + Opus 4.7
20%
OpenCode + GPT-5.5
100%
OpenCode + Sonnet 4.6
80%
OpenCode + Qwen3.7-Max
40%
OpenCode + GLM-5.2
60%
OpenCode + DeepSeek V4-Pro
40%
OpenCode + GLM-5.1
40%
OpenCode + Qwen 3.6 Plus
0%
OpenCode + MiniMax-M3
100%
OpenCode + MiniMax M2.7
0%
all checkers pass Type-A clean, Type-B fail fail checking timeout no result / incomplete
Strategy

What the paper asks for.

Scale the market excess return by the inverse of last month's realized variance (RV from sum of squared daily Mkt-RF), with a full-sample constant c chosen so the managed series matches unmanaged volatility: higher Sharpe and significant alpha vs. the unmanaged factor.

Type
technical
Implementation
pandas
Rebalance
monthly
Long-short
no
Declared task setup
Data
Bundled public task data
Operational scope
The paper-defined methodological core is retained; the fixed instruction instantiates only the data mappings, interfaces, and evaluation conventions needed for execution.
Six-stage quantitative strategy pipeline
  1. 01 Data loading
  2. 02 Feature engineering
  3. 03 Model fitting
  4. 04 Signal generation
  5. 05 Portfolio construction
  6. 06 Execution
Checker suite

18 properties · 80 attempts.

Each row is one requirement the strategy must satisfy. The bar reports the share of observed checker verdicts that pass; missing checker output is not converted into a property verdict.

TYPE A · DOMAIN-LEVEL (6)
  • A2
    no global normalize
    stage · signal gen
    74/80
  • A3
    no future data access
    stage · feature eng
    74/80
  • A4
    causal consistency
    stage · signal gen
    73/80
  • A7
    data frequency alignment
    stage · data loading
    72/80
  • A9
    signal delay test
    stage · execution
    73/80
  • A10
    e2e metric consistency
    stage · evaluation
    79/80
TYPE B · PAPER-SPECIFIC (12)
  • B1
    rv is sum of squared daily excess returns within calendar month
    §§2.2 “we keep the portfolio construction even simpler by using the previous month realized variance as a proxy for the conditional variance”
    provenance · paper-explicit
    72/80
  • B2
    weight uses previous month realized variance
    §§2.2 “previous month realized variance as a proxy for the conditional variance”
    provenance · paper-explicit
    74/80
  • B3
    managed factor return is weight times factor
    §§2.1 eq.(1) “f^σ_{t+1} = (c / σ̂²_t(f)) · f_{t+1}”
    provenance · paper-explicit
    73/80
  • B4
    constant c normalizes unconditional volatility to buy and hold
    §§2.1 paragraph after eq.(1) “we choose c so that the managed portfolio has the same unconditional standard deviation as the buy-and-hold portfolio.”
    provenance · paper-explicit
    73/80
  • B5
    monthly rebalance frequency
    §§2.1 eq.(1) “Each month our strategy increases or decreases risk exposure”
    provenance · paper-explicit
    74/80
  • B6
    linear scaling no thresholding or long short signal
    §§2.2 “we approximate the conditional risk-return trade-off by the inverse of the conditional variance”
    provenance · paper-explicit
    74/80
  • B7
    alpha from timeseries regression on original factor
    §§2.3 “We run a time-series regression of the volatility managed portfolio on the original factors”
    provenance · paper-explicit
    75/80
  • B8
    appraisal ratio is alpha over residual std
    §§2.3 “the managed factor excess Sharpe ratio (or 'appraisal ratio') given by α/σ_ε”
    provenance · paper-explicit
    75/80
  • B9
    factor used is market excess return mktrf
    §§2.4 / Table 1 “We first apply our analysis factor by factor. ... Mkt”
    provenance · paper-explicit
    74/80
  • B10
    annualized sharpe uses sqrt 12
    §Table 1 caption “annualized Sharpe ratio of the original MVE portfolio and the appraisal ratio”
    provenance · paper-explicit
    54/80
  • B11
    first month managed return is undefined due to lag
    §§2.2 “previous month realized variance”
    provenance · paper-explicit
    73/80
  • B12
    no position caps or leverage constraints in baseline
    §§2.2 “An appealing feature of this approach is that it can be easily implemented by an investor in real time”
    provenance · paper-explicit
    80/80
What this task reveals

The lowest-passing observed requirement is annualized sharpe uses sqrt 12: 54 of 80 observed checker verdicts pass. The full property list above shows whether errors concentrate in domain-level validity or the paper-specific strategy logic.

Open paper → All tasks
Next in Volatility Management →
T17 · Betting Against Beta