Post-Modern Portfolio Theory (PMPT)

PMPT replaces standard deviation with downside deviation, which only penalizes returns falling below a user-defined Minimum Acceptable Return (MAR). This corrects Modern Portfolio Theory's (MPT) flawed assumption that investors dislike all volatility equally.

• Formula: The square root of the probability-weighted squared deviations below the MAR.
• Asymmetry: Upside volatility (gains) is treated as beneficial, not risky.
• Realism: Aligns with behavioral finance findings that investors are loss-averse, not volatility-averse.

The Sortino Ratio is the PMPT analog to the Sharpe Ratio, measuring excess return per unit of bad risk. It is calculated as:

(Portfolio Return - MAR) / Downside Deviation

• Interpretation: A higher Sortino Ratio indicates more efficient compensation for taking on downside risk.
• Advantage: Unlike the Sharpe Ratio, it does not penalize a manager for generating large, positive outlier returns.
• Use Case: Preferred by hedge funds and CTAs whose return distributions are often non-normal and positively skewed.

The MAR is the foundational benchmark in PMPT, representing the return an investor must achieve to meet a specific financial goal. It replaces the risk-free rate as the dividing line between 'good' and 'bad' risk.

• Customization: Can be set to 0% (preservation of capital), an actuarial rate (pension liabilities), or an inflation-adjusted target.
• Liability-Driven: Directly links portfolio construction to the investor's future obligations.
• Dynamic: The MAR can be adjusted over time as an investor's funding status or goals change.

PMPT explicitly models the skewness and kurtosis of asset returns, rejecting MPT's assumption of a normal distribution. Financial returns exhibit 'fat tails' and asymmetry.

• Positive Skewness: A desirable trait where large positive returns are more frequent than large negative ones.
• Fat Tails: PMPT accounts for extreme events that standard deviation underestimates.
• Omega Ratio: A PMPT metric that captures the entire distribution by calculating the probability-weighted ratio of gains versus losses relative to a threshold.

Under MPT, the efficient frontier is plotted as expected return vs. standard deviation. Under PMPT, it is plotted as expected return vs. downside deviation.

• Visual Shift: The PMPT frontier often lies above and to the left of the MPT frontier, revealing that portfolios are more 'efficient' when only penalized for downside risk.
• Optimization: A PMPT optimizer will naturally tilt allocations toward assets with positive skewness and away from those with severe downside tail risk.
• Practical Outcome: Leads to portfolios that historically experience shallower drawdowns during market crises.

PMPT is mathematically aligned with Prospect Theory, developed by Kahneman and Tversky, which demonstrates that the pain of a loss is psychologically about twice as powerful as the pleasure of an equivalent gain.

• Loss Aversion: PMPT's focus on downside deviation directly models this psychological bias.
• Reference Dependence: The MAR serves as the 'reference point' from which gains and losses are evaluated.
• Practical Alignment: By matching the mathematical objective function to actual investor psychology, PMPT creates portfolios that investors are more likely to stick with during volatility.

The core risk measure in PMPT, replacing standard deviation. It calculates the dispersion of returns that fall below a user-defined Minimum Acceptable Return (MAR) , ignoring upside volatility entirely.

• Formula: Square root of the average squared deviations below MAR
• Contrast: Standard deviation penalizes positive returns; downside deviation does not
• Example: A portfolio with frequent large gains and small losses will have a low downside deviation but a high standard deviation

A risk-adjusted return metric developed by Frank Sortino that uses downside deviation in the denominator instead of standard deviation. It measures excess return per unit of bad risk.

• Formula: (Rp - MAR) / Downside Deviation
• Interpretation: Higher values indicate better risk-adjusted performance
• Advantage: Differentiates between harmful volatility and beneficial upside volatility
• Use Case: Evaluating hedge fund managers who target absolute returns above a specific hurdle rate

The threshold return below which an investor considers outcomes undesirable. MAR is the fulcrum that separates harmful downside risk from beneficial upside volatility in PMPT.

• Common MARs: Zero (preservation of capital), risk-free rate, actuarial required rate, or inflation rate
• Impact: Changing the MAR fundamentally alters the composition of the efficient frontier
• Example: A pension fund with a 7% actuarial assumption sets MAR at 7%; returns below this create a funding deficit

The set of optimal portfolios that maximize expected return for a given level of downside risk. Unlike the traditional efficient frontier, this surface shifts based on the investor's MAR.

• Construction: Uses downside deviation on the x-axis instead of standard deviation
• Shape: Often produces different optimal allocations than MVO, particularly for asymmetric return distributions
• Implication: Portfolios with positive skew (lottery-like payoffs) appear more attractive under PMPT than under MVO

PMPT explicitly accounts for the fact that real-world asset returns are not normally distributed. It handles skewness and kurtosis naturally, unlike MVO which assumes symmetry.

• Positive Skew: More frequent small losses and occasional large gains (e.g., venture capital, options strategies)
• Negative Skew: More frequent small gains and occasional large losses (e.g., carry trades, short volatility)
• PMPT Advantage: Correctly identifies positively skewed assets as less risky than MVO would suggest

A performance measure that considers the entire return distribution by calculating the probability-weighted ratio of gains versus losses relative to a threshold. It is a natural complement to PMPT.

• Formula: Integral of [1 - F(x)] dx above threshold divided by integral of F(x) dx below threshold
• Key Property: Incorporates all moments of the distribution (mean, variance, skewness, kurtosis)
• Decision Rule: Prefer the portfolio with the highest Omega ratio for a given threshold
• Link to PMPT: Both frameworks reject variance as a symmetric risk measure