paper_id	claim	claim_status	evidence_status	source_depth	source_quote	page_or_section	taxonomy_fit	audit_status
2309.08800v1	We introduce a computationally scalable framework for lead-lag detection in high-dimensional time series based on DTW, with clustering used as a denoising step.	supported	has evidence row	full-text	The remainder of the data pre-processing is the same as above. 6.3 Benchmark In order to evaluate our proposed methodology, we also introduce a benchmark to detect lead-lag relationships without the use of clustering.	16	in-scope: LLM extractor confirmed direction match	needs work; full-text verified; report=audit_report.md
2206.10173v1	Deriving an asymptotic statistical test to compare the amount of transfer entropy from one time series to another.	supported	has evidence row	full-text	Note that both the asymptotic analytical and the bootstraps results agree. 4.2 Comparing TEs Benchmarking the true null hypothesis, in this case, is much harder.	1	in-scope: LLM extractor confirmed direction match	needs work; full-text verified; report=audit_report.md
2601.01871v1	The proposed method delivers superior numerical performance compared to existing methods.	supported	has evidence row	full-text	Remark 3.3(Relation to mode estimation).Our formulation of the lead-lag parameter estimation problem is naturally connected to mode estimation for a probability density function, once the CPCF is replaced by the density.	Section 4	out-of-scope: LLM extractor flagged paper as not addressing direction	needs work; full-text verified; report=audit_report.md
2002.00208v3	We prove that our definitions and the proposed inference methods can address the arbitrary-time-lag influence between cause and effect.	supported	has evidence row	full-text	Lastly, the transfer entropy methods with bootstrapping almost failed to detect anything. /T_his is due to the weak signal of causal relations in real-world datasets. 9.4 Variable lags vs. fixed lag 9.4.1 VL-Granger causality.	1	out-of-scope: LLM extractor flagged paper as not addressing direction	needs work; full-text verified; report=audit_report.md
2305.06704v3	We introduce a computationally scalable pipeline for the robust detection of lead-lag relationships in high-dimensional time series.	supported	has evidence row	full-text	The remainder of the data pre-processing is the same as above. 6.3 Benchmark In order to evaluate our proposed methodology, we also introduce a benchmark to detect lead-lag relationships without the use of clustering.	4	in-scope: LLM extractor confirmed direction match	needs work; full-text verified; report=audit_report.md