Stock Return Prediction Using Television Advertising Data
American Journal of Undergraduate Research · 12.04.2026 · 23 (1) · 83
DOI: 10.33697/ajur.2026.169 ↗Kurz gefasst
Fachartikel in der Zeitschrift American Journal of Undergraduate Research, online erschienen am 12.04.2026. Verfasst von Haruto Hirata.
Abstract (Originalsprache)
This study proposes the TV Index, a novel alternative data source constructed from 6.6 million television commercial records spanning 965 Tokyo Stock Exchange-listed firms, and demonstrates its utility for systematic equity investment in the Japanese market. By applying Golden Cross strategies ( traditionally applied to price series) to this attention-based signal, this study shows that television advertising intensity contains predictive information about future stock returns that is not captured by conventional risk factors. The best-performing strategy achieves a Sharpe ratio of 1.030 and a cumulative return exceeding the TOPIX benchmark by approximately 65 percentage points over the 2019–2024 sample period. Multiple performance evaluations, including risk-adjusted analysis and sub-period decomposition, consistently confirm the robustness of these findings. These results demonstrate that television advertising data constitute a viable and practically implementable alternative data source, bridging behavioral finance theory with quantitative and econometric implementation. KEYWORDS: Stock Return Prediction; Television Advertising; TV Index; Alternative Data; Investor Attention; Golden Cross Strategy; Japanese Equity Market; Fama–French Model
Zitieren
APA
Hirata, H. (2026). Stock Return Prediction Using Television Advertising Data. American Journal of Undergraduate Research, 23(1), 83. https://doi.org/10.33697/ajur.2026.169
Vancouver
Hirata H. Stock Return Prediction Using Television Advertising Data. American Journal of Undergraduate Research. 2026;23(1):83. doi:10.33697/ajur.2026.169
BibTeX
@article{Hirata2026,
author = {Hirata, Haruto},
title = {Stock Return Prediction Using Television Advertising Data},
journal = {American Journal of Undergraduate Research},
year = {2026},
volume = {23},
number = {1},
pages = {83},
doi = {10.33697/ajur.2026.169}
}
EUVIMED