Paper: arXiv 2307.09392
Authors: Umut Cetin, Kasper Larsen
Abstract
In the dynamic discrete-time trading setting of Kyle (1985), we prove that Kyle’s equilibrium model is stable when there are one or two trading times. For three or more trading times, we prove that Kyle’s equilibrium is not stable. These theoretical results are proven to hold irrespectively of all Kyle’s input parameters.
Complexity vs Empirical Score
- Math Complexity: 9.5/10
- Empirical Rigor: 1.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper is heavily theoretical, proving stability properties of a canonical microstructure model using advanced functional analysis and fixed-point theory, with no empirical data or backtesting involved. Its results are purely analytical and abstract, making it typical of theoretical finance research.
Research Flowchart
flowchart TD
A["Research Goal:<br>Is Kyle's equilibrium stable?"] --> B["Methodology:<br>Mathematical Proofs"]
B --> C["Input Parameters:<br>Kyle's model assumptions"]
C --> D["Analysis of Trading Times:<br>N = 1, 2, 3, ..."]
D --> E{"Stable?"}
E -- N = 1, 2 --> F["Outcome: STABLE"]
E -- N >= 3 --> G["Outcome: NOT STABLE"]
F --> H["Conclusion:<br>Stability depends on trading periods"]
G --> H
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 proves that Kyle's equilibrium is stable for one or two trading times and not stable for three or more trading times, independent of all input parameters.
Source sentence
First, when N ∈{1, 2}, we prove that Kyle’s equilibrium is always stable. Second, when N ≥3, we prove that Kyle’s equilibrium is always not stable.
Assumes 6
- Noise traders' orders are Gaussian with mean zero and variance σ_u^2 Δ. model
Source sentence
The noise traders’ orders ∆un at trading time n ∈{1, ..., N} are Gaussian random variables with mean zero and variance σ2 u∆, where ∆> 0 is the time step.
- The stock's liquidating value is Gaussian with mean zero and variance Σ_0 > 0. model
Source sentence
The stock’s liquidating value is denoted by ˜v, which is assumed Gaussian with mean zero and variance Σ0 := V[˜v] > 0.
- Exogenous random variables (v~, Δu_1,…,Δu_N) are mutually independent. model
Source sentence
These exogenous random variables (˜v, ∆u1, ..., ∆uN) are assumed mutually independent.
- The insider's order at time 1 is measurable with respect to σ(v~). model
Source sentence
At time n = 1, the insider submits orders ∆x1 to the market makers.
Show 2 more
- For later times n≥2, the insider's order is measurable with respect to σ(v~, Δu_1,…,Δu_{n-1}). model
Source sentence
At later times n ∈{2, ..., N}, the insider submits orders ∆xn, which are required to be measurable with respect to σ(˜v, ∆u1, ..., ∆un−1).
- The model is discrete‑time with N ∈ ℕ trading times. model
Source sentence
2 Kyle’s discrete-time model This section briefly recalls the discrete-time model in Kyle (1985) with N ∈N trading times.
Data
No dataset named in the text.
Costs
Not applicable
Finds 5
- For N = 1, Kyle’s equilibrium is a super‑attractive stable fixed point, i.e., local convergence is faster than linear. theoretical
Source sentence
Even stronger, for N = 1, we show that Kyle’s equilibrium is a super-attractive fixed point in the sense that local convergence is strictly faster than linear.
- For N = 2, Kyle’s equilibrium is an attractive stable fixed point. theoretical
Source sentence
First, when N ∈{1, 2}, we prove that Kyle’s equilibrium is always stable.
- For N ≥ 3, Kyle’s equilibrium is a repelling (not stable) fixed point. theoretical
Source sentence
Second, when N ≥3, we prove that Kyle’s equilibrium is always not stable. Mathematically speaking, for N = 3, Kyle’s equilibrium⃗ˆβ is a repelling fixed point for the policy iteration operator T whereas the fixed point⃗β(∞) is an attractive fixed point for T. 10 5.2 Theory To rigorously disprove local stability when N ≥3, we iterate only in the third-to-last variable βN−2 whereas all other coefficients are set equal
- Even though not stable for N ≥ 3, the equilibrium is not unstable: perturbations of the last coordinate still converge back to the equilibrium. theoretical
Source sentence
For N ≥3, our next and last theoretical result shows that while Kyle’s equilibrium is not stable, it is also not unstable. Theorem 5.5. Let N ≥3. For ∆> 0, σu > 0, and Σ0 > 0, Kyle’s equilibrium is not unstable with respect to policy iterations for the insider in the sense of Definition 5.4.
- Numerical illustrations confirm non‑stability for N = 3 when the insider’s noise‑trader variance estimate is slightly misspecified. theoretical
Source sentence
We numerically compute the limit of the policy iterations when the insider’s measurement of noise-trader variance slightly differ from the correct noise-trader variance. Our numerics are based on a 10th digit difference.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 18 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.