Asymmetric multifractality and the evolution of market efficiency in Nepal: Regime-conditional evidence from NEPSE sectoral indices
Market efficiency in frontier markets is neither time-invariant nor uniform across sectors, yet most evidence assumes both. We analyse the NEPSE index and seven sectoral indices, July 2003 to December 2025, using asymmetric multifractal detrended fluctuation analysis within dated regimes. A heteroskedasticity-robust cumulative-sum-of-squares test dates breaks in five of eight series. Directional scaling is separated by the local trend, each difference carries a moving-block bootstrap interval, and shuffled and IAAFT surrogates isolate correlation- from distribution-driven multifractality. Generalized Hurst exponents exceed 0.5 in 11 of 12 series-regime cells and significantly exceed their shuffled benchmark in nine before multiplicity adjustment, and in three under Holm’s correction, though no bootstrap interval excludes 0.5, but only seven have spectra wider than their shuffled counterparts; the rest reflect fat tails, not dependence. Directional asymmetry is unsupported: downside spectra are wider in nine cells, but the sign test is insignificant (p = 0.146) and no cell is significant under both constructions. The spectrum narrows significantly across the break for NEPSE, Commercial Bank and Hotel, and the deficiency measure falls for Finance, against the presumption that frontier-market conditions deteriorate; Commercial Bank’s width narrowing is significant under both constructions at only two of five block lengths. A narrower spectrum is a change in scaling properties, not evidence of improved informational efficiency: Hotel and Finance are indistinguishable from their shuffled benchmarks in both regimes, and h(2) approaches 0.5 in no series. The paper contributes a regime-conditional asymmetric design, within-market heterogeneity, and a validation protocol that alters which conclusions survive.
Authors
- Sandip Paudel (ORCID: https://orcid.org/0000-0003-4689-7477)
- Pitambar Sapkota (ORCID: https://orcid.org/0009-0000-5160-9252)
Institutions
- Deen Dayal Upadhyaya Gorakhpur University (IN)
Publication Details
- Journal
- Cogent Economics & Finance
- Published
- 2026-11-18
- DOI
- https://doi.org/10.1080/23322039.2026.2761796
- Primary Topic
- Complex Systems and Time Series Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00