Peer-reviewed journal article Rajendran, S., & Singaravelu, D. (2026). Order-flow toxicity in cryptocurrency perpetual futures: Order flow signals, price-impact persistence, and the limits of volatility-based metrics.
International Review of Economics & Finance, 112, 105876 (Elsevier). Open access, CC BY 4.0.
First and corresponding author. Published online September 30, 2026.
doi:10.1016/j.iref.2026.105876
Why it matters for your system. Statistical validation of signals before anyone acts on them: the same standard we apply to numbers produced by financial data assistants. Financial data assistant →
Preprint (not peer reviewed) Rajendran, S., & Singaravelu, D. (2026). Evidence-bound factual repair in retrieval-augmented LLM answers: Separating semantic localization from deterministic realization.
Preprints.org, 202609.0490 (v1). CC BY 4.0.
First and corresponding author. Posted September 7, 2026; 1,300+ views and 300+ downloads by October 10, 2026.
doi:10.20944/preprints202609.0490.v1
Why it matters for your system. How to keep each sentence of a generated answer tied to its evidence, and repair the ones that drift. This is the method behind our assistants that cite or refuse. Assistant that cites or refuses →
Working paper Rajendran, S., & Singaravelu, D. (2026). Order-flow toxicity in cryptocurrency perpetual futures (working-paper version of the article above).
SSRN Working Paper No. 6344338.
Circulated in two versions from March 2026; 500+ downloads combined.
ssrn.com/abstract=6344338