Paper: arXiv 2310.12333

Authors: Jungjun Park, Andrew L. Nguyen

Abstract

In this paper, we study the Black-Litterman (BL) asset allocation model (Black and Litterman, 1990) under the hidden truncation skew-normal distribution (Arnold and Beaver, 2000). In particular, when returns are assumed to follow this skew normal distribution, we show that the posterior returns, after incorporating views, are also skew normal. By using Simaan three moments risk model (Simaan, 1993), we could then obtain the optimal portfolio. Empirical data show that the optimal portfolio obtained this way has less risk compared to an optimal portfolio of the classical BL model and that they become more negatively skewed as the expected returns of portfolios increase, which suggests that the investors trade a negative skewness for a higher expected return. We also observe a negative relation between portfolio volatility and portfolio skewness. This observation suggests that investors may be making a trade-off, opting for lower volatility in exchange for higher skewness, or vice versa. This trade-off indicates that stocks with significant price declines tend to exhibit increased volatility.

Complexity vs Empirical Score

  • Math Complexity: 8.0/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper involves advanced mathematics including hidden truncation skew-normal distributions, Bayesian posterior derivations, and Simaan’s three-moment optimization framework, but the empirical evidence is limited to described observations without backtesting code, specific datasets, or implementation details.

Research Flowchart

  flowchart TD
  A["Research Goal: Study BL Model under<br>Hidden Truncation Skew-Normal Distribution"] --> B["Data: Historical Asset Returns"]
  B --> C["Methodology: Parametric Assumptions<br>Skew-Normal Distribution for Returns"]
  C --> D["Process: Derive Posterior Returns<br>Skew-Normal Distribution"]
  D --> E["Process: Optimize Portfolio using<br>Simaan Three-Moments Risk Model"]
  E --> F{"Key Findings & Outcomes"}
  F --> G["Optimal Portfolio: Lower Risk<br>vs. Classical BL Model"]
  F --> H["Trade-off: Higher Expected Return<br>leads to More Negative Skewness"]
  F --> I["Observation: Negative Relation<br>between Volatility and Skewness"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper extends the Black‑Litterman model to a hidden truncation skew‑normal distribution, allowing for skewness and heavier tails in asset returns.

Source sentence
In this paper, we contribute to extend the BL model to the hidden truncation model (distribution) by departing from the normal market assumption.

Assumes 3

  • Asset returns are assumed to follow the hidden truncation skew‑normal distribution. statistical
    Source sentence
    when returns are assumed to follow this skew normal distribution, we show that the posterior returns, after incorporating views, are also skew normal.
  • Investor views are modeled as normally distributed with a specified confidence matrix. model
    Source sentence
    ... And the following matrix gives a measure of the investors’ confidence level on the views.
  • The hidden truncation skew‑normal distribution provides heavier tails and higher peaks than the ordinary normal distribution. statistical
    Source sentence
    This class has the advantage of allowing for heavier tails and higher peaks than the ordinary normal distribution.

Data

  • 13 equities sample market-prices unknown
    AAPL, AMZN, GOOGL, MSFT, XOM, … (13 equities)
    Source sentence
    Table 5: Maximum likelihood estimates of µ, λ1, and τ 1. Ticker ˆµ ˆλ1 ˆτ 1 AAPL 0.113 –1.918 –1.783 AMZN 0.077 –0.648 –0.600 GOOGL 0.074 –0.938 –0.864 MSFT 0.063 –0.889 –0.817 XOM 0.043 –0.715 –0.658

Costs

Not modeled A trading or execution result is reported without costs.

Finds 4

  • The optimal portfolio derived from the hidden truncation skew‑normal BL model has lower risk than the optimal portfolio from the classical BL model. performance
    Source sentence
    Empirical data show that the optimal portfolio obtained this way has less risk compared to a
  • Likelihood‑ratio test strongly rejects normality, supporting the skew‑normal assumption for the data (p‑value = 0.0015). statistical
    Source sentence
    Observed–data Log-likelihood normal 3652.19 Log-likelihood skew-normal 3668.80 2 × ln(LR) 33.212 P-value 0.0015
  • There is a negative relation between portfolio expected return and portfolio skewness across different N values. negative
    Source sentence
    We find that there is a negative relation between portfolio expected return and portfolio skewness given the values of N.
  • Sharpe ratios of skew‑BL optimal portfolios are reported for various target returns, generally higher than the standard BL portfolios. performance
    Source sentence
    Table 9: Sharpe Ratio. This table reports Sharpe ratios for BL and skBL optimal port‑folios.

Checked by others

Supported by

  • Phase Transitions in Financial Markets Using the Ising Model: A Statistical Mechanics Perspective (2025)
    On the question: Do asset return distributions deviate from normality, showing skewness and heavy tails?
    Compare the findings

    This paper: Likelihood‑ratio test strongly rejects normality, supporting the skew‑normal assumption for the data (p‑value = 0.0015).

    That paper: Both the Ising model returns and the S&P 500 returns display significant skewness and leptokurtic (heavy‑tailed) distributions, rejecting normality in Shapiro‑Wilk and Jarque‑Bera tests.

    Finding A rejects normality (p=0.0015) favoring skew‑normal; Finding B also rejects normality via Shapiro‑Wilk and Jarque‑Bera, showing skewness and leptokurtosis.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.