Paper: arXiv 2110.07138

Authors: Zura Kakushadze, Willie Yu

Abstract

We discuss how to build ETF risk models. Our approach anchors on i) first building a multilevel (non-)binary classification/taxonomy for ETFs, which is utilized in order to define the risk factors, and ii) then building the risk models based on these risk factors by utilizing the heterotic risk model construction of https://ssrn.com/abstract=2600798 (for binary classifications) or general risk model construction of https://ssrn.com/abstract=2722093 (for non-binary classifications). We discuss how to build an ETF taxonomy using ETF constituent data. A multilevel ETF taxonomy can also be constructed by appropriately augmenting and expanding well-built and granular third-party single-level ETF groupings.

Complexity vs Empirical Score

  • Math Complexity: 7.0/10
  • Empirical Rigor: 4.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper introduces advanced mathematical frameworks like heterotic risk models and discusses factor covariance matrix construction with detailed formulas, but lacks concrete backtesting results, performance metrics, or implementation-specific data.

Research Flowchart

  flowchart TD
  A["Research Goal: Build ETF Risk Models"] --> B{"Select Classification Type"}
  B -->|"Binary"| C1["Heterotic Risk Model"]
  B -->|"Non-Binary"| C2["General Risk Model"]
  
  D["ETF Constituent Data"] --> E["Construct Multilevel ETF Taxonomy"]
  E --> B
  
  C1 --> F["Risk Factor Definition"]
  C2 --> F
  
  F --> G["Compute ETF Risk Models"]
  G --> H["Key Findings: Improved Risk Measurement & Taxonomy Framework"]

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 building ETF risk models by first creating a multilevel ETF taxonomy to define risk factors and then constructing heterotic (or general) risk models based on these factors.

Source sentence
Our approach anchors on i) first building a multilevel (non-)binary classification/taxonomy for ETFs, which is utilized in order to define the risk factors, and ii) then building the risk models based on these risk factors by utilizing the heterotic risk model construction of https://ssrn.com/abstract=2600798 (for binary classifications) or general risk model construction of https://ssrn.com/abstract=2722093 (for

Assumes 4

  • A 2‑level classification (granular level and asset‑class level) is sufficient for the heterotic risk model construction. model
    Source sentence
    In fact, for our purposes here (i.e., for ensuring that the factor covariance matrix in the heterotic risk model construction is nonsingular – see above) a 2-level classification will suffice.
  • The first, granular level classification is binary, i.e., each ETF belongs to exactly one category. model
    Source sentence
    Let us assume that the first, granular level is binary (i.e., each ETF is classified into one and only one category).
  • The second, asset‑class level classification is also binary. model
    Source sentence
    By definition, so is the second level as asset class assignments are binary (assuming the asset class categories include the multi-asset category – see above).
  • ETF returns can be linearly decomposed into contributions from a set of common factors plus idiosyncratic noise. model
    Source sentence
    So, just as in the case of stocks, we can forecast the ETF covariance matrix via a multifactor risk model, where ETF returns are (linearly) decomposed into contributions stemming from some number K of common underlying factors plus idiosyncratic “noise” pertaining to each ETF individually.

Data

  • ETF constituent data fundamentals unknown
    Source sentence
    We discuss how to build an ETF taxonomy using ETF constituent data.
  • BICS (Bloomberg Industry Classification System) · Bloomberg fundamentals unknown
    Source sentence
    11 E.g., BICS (Bloomberg Industry Classification System), GICS (Global Industry Classification Standard), ICB (Industry Classification Benchmark), SIC (Standard Industrial Classification), etc.

Costs

Not applicable The paper reports no trading or execution result.

Finds

No quantified findings extracted.

Checked by others

No other paper in the archive has been found to support or contradict these findings. Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-17 · same card as JSON · A language model chose these items: check the source sentence before relying on one.