Paper: arXiv 2401.05423

Authors: Victor Ujaldon Garcia

Abstract

Four markets are considered: Cryptocurrencies / South American exchange rate / Spanish Banking indices and European Indices and studied using TDA (Topological Data Analysis) tools. These tools are used to predict and showcase both strengths and weakness of the current TDA tools. In this paper a new tool \(L0\) norm is defined and complemented with the already existing \(C1\) norm.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 4.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper introduces and applies advanced topological mathematics (TDA, homology, persistence landscapes) with dense theoretical exposition, but the empirical validation relies on visual inspection of a few specific market examples without statistical testing, out-of-sample validation, or code/implementation details.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>New TDA tools for market prediction"] --> B["Data Collection:<br>4 Markets: Crypto, Latam FX, Spanish Banks, EU Indices"]
  B --> C["Methodology:<br>Topological Data Analysis - TDA"]
  C --> D["Computational Process:<br>Define & Apply L0 norm<br>Complement with C1 norm"]
  D --> E["Key Findings:<br>L0 + C1 norms predict market trends<br>Reveals TDA strengths & weaknesses"]

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 defines a new L0 norm and complements it with the existing C1 norm for analyzing time series.

Source sentence
In this paper a new tool $L0$ norm is defined and complemented with the already existing $C1$ norm.

Assumes 3

  • The analysis uses closing values of stocks, indices, or currency exchange rates to construct the time series. data
    Source sentence
    To construct a time series {Xa}a∈(0,T) able to study market data we consider the closing values of the stock/index/currency exchange value of n units and compute the log scale.
  • Log returns are computed as log(x_{i+1}/x_i) for each series. model
    Source sentence
    That is: for a 1 dimensional data set {xi,j} we consider Xj = {log(xi+1,j/xi,j)} and the time value {Xa j } = {log(xi+1,j/xi, j)}i=a.
  • A standard time window of T = 30 and j = 4 units is used for all norm calculations. model
    Source sentence
    Finally we consider the temporal cloud point {Xa}a∈(0,T) = {Xa 1, ..., Xa n}a∈(0,T) and we study the L0, L1, C1 norms with a time window T. In every case we used the standard time window T = 30 and j = 4 units.

Data

  • Cryptocurrencies, South American exchange rate, Spanish Banking indices, European Indices market-prices unknown
    Source sentence
    Four markets are considered: Cryptocurrencies / South American exchange rate / Spanish Banking indices and European Indices and studied using TDA (Topological Data Analysis) tools.

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • L1 norm has great advantages for time series analysis but cannot capture all topological information in the point cloud. theoretical
    Source sentence
    5 Conclusion It is clear that L1 norm has great advantages when it comes to time series analysis but it’s unable to unravel all the topological information hidden in the point cloud.
  • L0 and C1 norms can complement L1 to determine market stability and identify critical transition states. theoretical
    Source sentence
    That’s where L0 and C1 can complement the initial information of L1 to determine if a market is stable and whether the market is in a critical transition state.
  • Four examples are showcased demonstrating the usefulness of each norm. other
    Source sentence
    Here we have showcased four examples and how every norm is useful.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 10 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.