Kalimati Vegetable Price Index Forecasting with a Momentum Corrected Online Stacking Ensemble
Sahaj Raj Malla · 2026
Read Kalimati Vegetable Price Index Forecasting with a Momentum Corrected Online Stacking Ensemble by Sahaj Raj Malla (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-30720.txt.
Description
Malla's 2026 study introduces the Kalimati Vegetable Price Index, a composite of 135 daily commodities from Kathmandu over ten years. The index creates a stable broad signal, cutting single-crop noise. From 64 features, fourteen models were tested. The models span statistical, tree-based, deep learning, hybrid, and transformer designs, across horizons of 7 to 90 days. Tree-based ensembles proved most robust, while statistical models and transformers struggled with noise. A momentum-corrected stacking ensemble performed best at 90 days, with low percentage error and most variance explained.
Catalog record
| CREATOR | Sahaj Raj Malla |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40444 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.30720