Files

Tim Olson 47471b7700 Expand model tag support: add GLM-5.1, simplify Anthropic IDs, scan tags anywhere in message

- Flink update_bars debouncing
- update_bars subscription idempotency bugfix
- Price decimal correction bugfix of previous commit
- Add GLM-5.1 model tag alongside renamed GLM-5
- Use short Anthropic model IDs (sonnet/haiku/opus) instead of full version strings
- Allow @tags anywhere in message content, not just at start
- Return hasOtherContent flag instead of trimmed rest string
- Only trigger greeting stream when tag has no other content
- Update workspace knowledge base references to platform/workspace and platform/shapes
- Hierarchical knowledge base catalog
- 151 Trading Strategies knowledge base articles
- Shapes knowledge base article
- MutateShapes tool instead of workspace patch

2026-04-28 15:05:15 -04:00

4.0 KiB

Raw Blame History

description, tags

description

Distress Risk Puzzle

Section: 15.3 | Asset Class: Distressed Assets (Equities) | Type: Factor / Long-Short

Overview

Early studies suggested that companies more prone to bankruptcy offer higher returns as a risk premium. However, more recent and robust studies find the opposite: distressed (high bankruptcy probability) companies do not outperform healthier ones, and healthier companies actually offer higher returns. This is the "distress risk puzzle." The strategy exploits it by buying the safest companies and selling the riskiest (a healthy-minus-distressed, or HMD, zero-cost long-short portfolio).

Construction / Mechanics

Estimate bankruptcy probability P_i for each stock i = 1, ..., N using, e.g., logistic regression on financial variables:
```
logit(P_i) = β_0 + β_1 × (leverage) + β_2 × (profitability) + ...
```
Sort stocks into deciles by P_i
Construct zero-cost portfolio:
- Short the top decile (highest P_i — most distressed)
- Long the bottom decile (lowest P_i — healthiest)
Rebalance monthly (annual rebalancing produces similar returns)

Return Profile

The HMD portfolio profits when healthy stocks outperform distressed stocks, which empirical evidence suggests happens persistently. Returns are driven by the cross-sectional spread in returns between the safest and riskiest firms. The strategy has equity-like volatility and is exposed to periods of market stress.

Key Parameters / Signals

Bankruptcy probability P_i: core signal; modeled via logistic regression or other classification methods on financial variables (leverage, profitability, coverage ratios, market-to-book, etc.)
Decile cutoffs: top and bottom decile are standard; tighter cutoffs increase signal strength but reduce breadth
Rebalancing frequency: monthly is standard; annual rebalancing yields similar returns with lower turnover

Variations

15.3.1 Distress Risk Puzzle — Risk Management

The standard HMD strategy has a high time-varying market beta that turns significantly negative following market downturns (associated with increased volatility). This can cause large losses when the market rebounds abruptly. To mitigate this, a volatility-scaled version is used:

HMD* = (σ_target / σ_hat) × HMD           (519)

HMD = standard HMD portfolio return (from Section 15.3)
σ_target = target volatility level, typically 10%–15% (set per trader preferences)
σ_hat = estimated realized volatility over the prior year using daily data

Interpretation:

If σ_hat = σ_target: 100% of the investment is allocated
If σ_hat > σ_target: allocation is reduced below 100% (de-leverage in high-vol regimes)
If σ_hat < σ_target: allocation exceeds 100% (leverage in low-vol regimes)

Alternatively, the allocation can be capped at 100% (min(σ_target / σ_hat, 1)) to avoid leverage entirely.

Notes

The distress risk puzzle is a well-documented anomaly but its persistence is debated; it may partly reflect data-mining or survivorship bias
The HMD strategy has high time-varying beta to the equity market; risk management via volatility scaling (15.3.1) is essential for production use
Bankruptcy probability models require regular recalibration; financial ratios used as inputs are sensitive to accounting changes
Short-selling the most distressed stocks can be expensive (high borrow costs) and difficult (low float, high short interest)
Regulatory restrictions on short selling may limit implementation in certain markets or during market stress periods
Similar time-varying beta behavior is observed in other factor-based strategies (momentum, value, etc.), suggesting a common structural risk

4.0 KiB Raw Blame History Unescape Escape