Paper: arXiv 2504.07929
Authors: Victor Olkhov
Abstract
The variance measures the portfolio risks the investors are taking. The investor, who holds his portfolio and doesn’t trade his shares, at the current time can use the time series of the market trades that were made during the averaging interval with the securities of his portfolio and assess the current return, variance, and hence the current risks of his portfolio. We show how the time series of trades with the securities of the portfolio determine the time series of trades with the portfolio as a single market security. The time series of trades with the portfolio determine its return and variance in the same form as the time series of trades with securities determine their returns and variances. The description of any portfolio and any single market security is equal. The time series of the portfolio trades define the decomposition of the portfolio variance by its securities, which is a quadratic form in the variables of relative amounts invested into securities. Its coefficients themselves are quadratic forms in the variables of relative numbers of shares of its securities. If one assumes that the volumes of all consecutive deals with each security are constant, the decomposition of the portfolio variance coincides with Markowitz’s (1952) variance, which ignores the effects of random trade volumes. The use of the variance that accounts for the randomness of trade volumes could help majors like BlackRock, JP Morgan, and the U.S. Fed to adjust their models, like Aladdin and Azimov, to the reality of random markets.
Complexity vs Empirical Score
- Math Complexity: 7.5/10
- Empirical Rigor: 3.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper employs advanced mathematical derivations, including quadratic forms and decomposition of portfolio variance based on time series of trades, indicating high mathematical density. However, it lacks empirical validation, backtests, or implementation details, relying solely on theoretical framework without data-driven evidence.
Research Flowchart
flowchart TD A["Research Goal: Model Portfolio Variance <br> Accounting for Random Trade Volumes"] --> B["Data Input: <br> Time Series of Market Trades <br> (Securities & Portfolio)"] B --> C["Methodology: <br> Treat Portfolio as Single Market Security"] C --> D["Computation: <br> Calculate Return & Variance <br> Using Trade Time Series"] D --> E["Decomposition: <br> Quadratic Form in <br> Relative Investments"] E --> F["Finding 1: <br> Standard Markowitz Variance <br> (when trade volumes are constant)"] E --> G["Finding 2: <br> New Variance Metric <br> (accounts for random trade volumes)"]
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 proposes a market‑based portfolio variance model that derives the portfolio return and variance from the time series of market trades with the portfolio’s securities, treating the portfolio as a single market security.
Source sentence
Market-based portfolio variance Θ(t,t0) that accounts for the randomness of the volumes of the consecutive trades with the securities of the portfolio changes the existing methods for optimal selection and portfolio compositions.
Assumes 2
- Volumes of all consecutive deals with each security are constant. model
Source sentence
If one assumes that the volumes of all consecutive deals with each security are constant, the decomposition of the portfolio variance coincides with Markowitz's (1952) variance, which ignores the effects of random trade volumes. If one assumes that the volumes of trades Uj(ti) with all securities j=1,..J of the portfolio during Δ (2.4) are constant Uj(ti)=Uj, then the volumes W(ti)=W of trade with the portfolio are
- Market trades with all securities occur simultaneously at the same time intervals with a constant short time span ε. execution
Source sentence
For convenience, we assume that market trades with all securities j=1,..J of the portfolio occur simultaneously at the same time ti with a short time span ε>0 between the trades and assume that ε is constant and the same for the trades with all securities j=1,..J. For each averaging time interval Δ,
Data
No dataset named in the text.
Costs
Not applicable
Finds 2
- If trade volumes are constant, the market‑based variance Θ(t,t0) reduces to Markowitz’s variance expression. theoretical
Source sentence
Finaly, obtain that if one assumes that all trade volumes Uj(ti)=Uj with all securities j=1,..J of the portfolio are constant, the variance Θ(t,t0) (B.21) of returns of the portfolio coincides with Markowitz’s expressions (1.2; 1.3) of the portfolio variance: 𝛩(𝑡, 𝑡0) = ∑ 𝜃𝑗𝑘(𝑡) 𝐽 𝑗,𝑘=1 𝑋𝑗(𝑡0)𝑋𝑘(𝑡0)
- The decomposition of the portfolio variance Θ(t,t0) is a quadratic form in the variables of relative amounts invested, with coefficients that are themselves quadratic forms, reflecting the impact of random trade volumes. theoretical
Source sentence
The decomposition of the variance Θ(t,t0) (4.7) is a quadratic form in variables of relative amounts Xj(t0) invested into securities with the coefficients jk(t,t0) (4.8), which, in their turn, are quadratic forms in variables xj(t0) (4.2; 4.4).
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 6 related papers). Coverage is partial, so this is neither replication nor refutation.
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.