Interpretable deep learning for modeling policy uncertainty and firm-specific risk: Evidence from advanced and emerging markets

Hyder Ali; Salma Naz; Hyder Ali; Salma Naz

doi:10.3934/DSFE.2026006

Data Science in Finance and Economics

2026, Volume 6, Issue 1: 147-182. doi: 10.3934/DSFE.2026006

Previous Article Next Article

Research article

Interpretable deep learning for modeling policy uncertainty and firm-specific risk: Evidence from advanced and emerging markets

Hyder Ali ^,,
Salma Naz

Faculty of Management Sciences, Sukkur IBA University, Sukkur 65200, Pakistan

Received: 15 July 2025 Revised: 19 November 2025 Accepted: 09 March 2026 Published: 18 March 2026
JEL Codes: C45, D81, E44, G12, G15, G32

This study examines how multidimensional policy-related uncertainty influences firm-specific risk by integrating econometric analysis with interpretable deep learning. Using daily stock-level data for Nikkei 225 firms from 2000–2023, we construct measures of idiosyncratic volatility (IVOL) and combine them with a multisource policy uncertainty index capturing seven domains: economic, fiscal, monetary, trade, exchange rate, energy-related, and geopolitical risk. To model nonlinear interactions and temporal persistence across uncertainty dimensions, we employ a long short-term memory (LSTM) autoencoder and benchmark its performance against principal component analysis (PCA) and kernel PCA (KPCA). The LSTM-based composite index exhibits the strongest explanatory power, showing that higher policy uncertainty systematically amplifies IVOL. Model transparency is ensured through SHapley Additive exPlanations (SHAP), which highlight fiscal, monetary, and geopolitical uncertainty as the dominant contributors. Heterogeneity analyses reveal that the sensitivity of IVOL to uncertainty varies by sector, firm size, and profitability. To evaluate external validity, we extend the analysis to several advanced and emerging markets, including the United States, Germany, South Korea, India, and Indonesia, and find a consistent uncertainty-IVOL relationship, with Japan displaying the highest sensitivity. The results underscore the value of combining deep learning with explainable artificial intelligence (AI) for financial risk assessment in uncertainty-driven environments.
- idiosyncratic volatility,
- policy uncertainty,
- interpretable machine learning,
- LSTM autoencoder,
- SHAP,
- nonlinear dimensionality reduction,
- financial risk forecasting,
- advanced and emerging markets
Citation: Hyder Ali, Salma Naz. Interpretable deep learning for modeling policy uncertainty and firm-specific risk: Evidence from advanced and emerging markets[J]. Data Science in Finance and Economics, 2026, 6(1): 147-182. doi: 10.3934/DSFE.2026006

Related Papers:

Abstract

This study examines how multidimensional policy-related uncertainty influences firm-specific risk by integrating econometric analysis with interpretable deep learning. Using daily stock-level data for Nikkei 225 firms from 2000–2023, we construct measures of idiosyncratic volatility (IVOL) and combine them with a multisource policy uncertainty index capturing seven domains: economic, fiscal, monetary, trade, exchange rate, energy-related, and geopolitical risk. To model nonlinear interactions and temporal persistence across uncertainty dimensions, we employ a long short-term memory (LSTM) autoencoder and benchmark its performance against principal component analysis (PCA) and kernel PCA (KPCA). The LSTM-based composite index exhibits the strongest explanatory power, showing that higher policy uncertainty systematically amplifies IVOL. Model transparency is ensured through SHapley Additive exPlanations (SHAP), which highlight fiscal, monetary, and geopolitical uncertainty as the dominant contributors. Heterogeneity analyses reveal that the sensitivity of IVOL to uncertainty varies by sector, firm size, and profitability. To evaluate external validity, we extend the analysis to several advanced and emerging markets, including the United States, Germany, South Korea, India, and Indonesia, and find a consistent uncertainty-IVOL relationship, with Japan displaying the highest sensitivity. The results underscore the value of combining deep learning with explainable artificial intelligence (AI) for financial risk assessment in uncertainty-driven environments.

References

[1]	Ang A, Hodrick RJ, Xing Y, et al. (2006) The cross‐section of volatility and expected returns. J Financ 61: 259–299. https://doi.org/10.1111/j.1540-6261.2006.00836.x doi: 10.1111/j.1540-6261.2006.00836.x
[2]	Ang A, Hodrick RJ, Xing Y, et al. (2009) High idiosyncratic volatility and low returns: International and further us evidence. J Financ Econ 91: 1–23. https://doi.org/10.1016/j.jfineco.2007.12.005 doi: 10.1016/j.jfineco.2007.12.005
[3]	Arbatli Saxegaard EC, Davis SJ, Ito A, et al. (2017) Policy uncertainty in Japan. IMF Working Pap 2017: 1–48. https://doi.org/10.5089/9781484300671.001 doi: 10.5089/9781484300671.001
[4]	Bahmani-Oskooee M, Nayeri MM (2020) Policy uncertainty and the demand for money in Japan. Rev Econ Anal 12: 73–87. https://doi.org/10.15353/rea.v12i1.1692 doi: 10.15353/rea.v12i1.1692
[5]	Baker SR, Bloom N, Davis SJ (2016) Measuring economic policy uncertainty. Q J Econ 131: 1593–1636. https://doi.org/10.1093/qje/qjw024 doi: 10.1093/qje/qjw024
[6]	Bali TG, Cakici N (2008) Idiosyncratic volatility and the cross section of expected returns. J Financ Quant Anal 43: 29–58. https://doi.org/10.1017/S002210900000274X doi: 10.1017/S002210900000274X
[7]	Bloom N (2009) The impact of uncertainty shocks. Econometrica 77: 623–685. https://doi.org/10.3982/ECTA6248 doi: 10.3982/ECTA6248
[8]	Bouri E, Gkillas K, Gupta R, et al. (2020) Monetary policy uncertainty and jumps in advanced equity markets. J Risk 23: 101–112. https://doi.org/10.21314/JOR.2020.441 doi: 10.21314/JOR.2020.441
[9]	Bradley M, Jarrell GA, Kim EH (1984) On the existence of an optimal capital structure: Theory and evidence. J Financ 39: 857–878. https://doi.org/10.1111/j.1540-6261.1984.tb03680.x doi: 10.1111/j.1540-6261.1984.tb03680.x
[10]	Caldara D, Iacoviello M (2022) Measuring geopolitical risk. Am Econ Rev 112: 1194–1225. https://doi.org/10.1257/aer.20191823 doi: 10.1257/aer.20191823
[11]	Caldara D, Iacoviello M, Molligo P, et al. (2020) The economic effects of trade policy uncertainty. J Monetary Econ 109: 38–59. https://doi.org/10.1016/j.jmoneco.2019.11.002 doi: 10.1016/j.jmoneco.2019.11.002
[12]	Cameron AC, Miller DL (2015) A practitioner's guide to cluster-robust inference. J Human Resour 50: 317–372. https://doi.org/10.3368/jhr.50.2.317 doi: 10.3368/jhr.50.2.317
[13]	Cao C, Simin T, Zhao J (2008) Can growth options explain the trend in idiosyncratic risk? Rev Financi Stud 21: 2599–2633. https://doi.org/10.1093/rfs/hhl039 doi: 10.1093/rfs/hhl039
[14]	Chen L, Zhao X (2006) On the relation between the market-to-book ratio, growth opportunity, and leverage ratio. Financ Res Lett 3: 253–266. https://doi.org/10.1016/j.frl.2006.06.003 doi: 10.1016/j.frl.2006.06.003
[15]	Chen X, Li Y, Xiao J, et al. (2020) Oil shocks, competition, and corporate investment: evidence from china. Energy Econ 89: 104819. https://doi.org/10.1016/j.eneco.2020.104819 doi: 10.1016/j.eneco.2020.104819
[16]	Dang THN, Nguyen CP, Lee GS, et al. (2023) Measuring the energy-related uncertainty index. Energy Econ 124: 106817. https://doi.org/10.1016/j.eneco.2023.106817 doi: 10.1016/j.eneco.2023.106817
[17]	Fama EF, French KR (1992) The cross‐section of expected stock returns. J Financ 47: 427–465. https://doi.org/10.1111/j.1540-6261.1992.tb04398.x doi: 10.1111/j.1540-6261.1992.tb04398.x
[18]	Fama EF, French KR (2015) A five-factor asset pricing model. J Financ Econ 116: 1–22. https://doi.org/10.1016/j.jfineco.2014.10.010 doi: 10.1016/j.jfineco.2014.10.010
[19]	Ferreira MA, Laux PA (2007) Corporate governance, idiosyncratic risk, and information flow. J Financ 62: 951–989. https://doi.org/10.1111/j.1540-6261.2007.01228.x doi: 10.1111/j.1540-6261.2007.01228.x
[20]	Frank MZ, Goyal VK (2009) Capital structure decisions: which factors are reliably important? Financ Manage 38: 1–37. https://doi.org/10.1111/j.1755-053X.2009.01026.x doi: 10.1111/j.1755-053X.2009.01026.x
[21]	Fu F (2009) Idiosyncratic risk and the cross-section of expected stock returns. J Financ Econ 91: 24–37. https://doi.org/10.1016/j.jfineco.2008.02.003 doi: 10.1016/j.jfineco.2008.02.003
[22]	Gaspar J, Massa M (2006) Idiosyncratic volatility and product market competition. J Bus 79: 3125–3152. https://doi.org/10.1086/505251 doi: 10.1086/505251
[23]	Ghosh S, Adebayo TS (2024) The effect of world policy uncertainty and geopolitical risk factors on export-led growth for japan: novel insights by wavelet local multiple correlation methods. Qual Quant 58: 3605–3633. https://doi.org/10.1007/s11135-023-01814-5 doi: 10.1007/s11135-023-01814-5
[24]	Gu S, Kelly B, Xiu D (2020) Empirical asset pricing via machine learning. Rev Financ Stud 33: 2223–2273. https://doi.org/10.1093/rfs/hhaa009 doi: 10.1093/rfs/hhaa009
[25]	Hasan MM, Habib A (2017) Firm life cycle and idiosyncratic volatility. Int Rev Financ Anal 50: 164–175. https://doi.org/10.1016/j.irfa.2017.01.003 doi: 10.1016/j.irfa.2017.01.003
[26]	Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9: 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735 doi: 10.1162/neco.1997.9.8.1735
[27]	Jiang F, Cai W, Wang X, et al. (2018) Multiple large shareholders and corporate investment: Evidence from china. J Corp Financ 50: 66–83. https://doi.org/10.1016/j.jcorpfin.2018.02.001 doi: 10.1016/j.jcorpfin.2018.02.001
[28]	Jiang GJ, Xu D, Yao T (2009) The information content of idiosyncratic volatility. J Financ Quant Anal 44: 1–28. https://doi.org/10.1017/S0022109009090073 doi: 10.1017/S0022109009090073
[29]	Jurado K, Ludvigson SC, Ng S (2015) Measuring uncertainty. Am Econ Rev 105: 1177–1216. https://doi.org/10.1257/aer.20131193 doi: 10.1257/aer.20131193
[30]	Kang W, de Gracia FP, Ratti RA (2017) Oil price shocks, policy uncertainty, and stock returns of oil and gas corporations. J Int Money Financ 70: 344–359. https://doi.org/10.1016/j.jimonfin.2016.10.003 doi: 10.1016/j.jimonfin.2016.10.003
[31]	Khan N, Zada H, Siddiqui O, et al. (2025) Sectoral response to economic policy uncertainty in japan: An empirical evidence from the cross-quantilogram approach. Comput Econ, 1–36. https://doi.org/10.1007/s10614-025-10867-7
[32]	Krol R (2014) Economic policy uncertainty and exchange rate volatility. Int Financ 17: 241–256. https://doi.org/10.1111/infi.12049 doi: 10.1111/infi.12049
[33]	Kumar S, Nor E, Azman NHN, et al. (2023) Commonality in liquidity: Evidence from emerging asian actively-managed equity funds. Asian Acad Manage J Account & Financ, 19. https://doi.org/10.21315/aamjaf2023.19.2.8
[34]	Li Y, Mu Y, Qin T (2021) Economic uncertainty: A key factor to understanding idiosyncratic volatility puzzle. Financ Res Lett 42: 101938. https://doi.org/10.1016/j.frl.2021.101938 doi: 10.1016/j.frl.2021.101938
[35]	Liang CYC, Tang Z (2018) Idiosyncratic return volatility, uncertainty, and asset pricing implications. Working paper, Available from: https://www.teaweb.org.tw/UpFile/1/Attach/2018-12/11122610998.pdf.
[36]	Lins KV, Servaes H, Tamayo A (2017) Social capital, trust, and firm performance: The value of corporate social responsibility during the financial crisis. J Financ 72: 1785–1824. https://doi.org/10.1111/jofi.12505 doi: 10.1111/jofi.12505
[37]	Liu R, Segnon M, Gupta R, et al. (2025) Conventional and unconventional monetary policy rate uncertainty and stock market volatility: A forecasting perspective. Stud Nonlinear Dyn Economet. https://doi.org/10.1515/snde-2024-0108
[38]	Liu S, Kong A, Gu R, et al. (2019) Does idiosyncratic volatility matter?—evidence from chinese stock market. Physica A 516: 393–401. https://doi.org/10.1016/j.physa.2018.09.184 doi: 10.1016/j.physa.2018.09.184
[39]	Liu X, Yang Q, Wei K, et al. (2024) ESG rating disagreement and idiosyncratic return volatility: Evidence from china. Res Int Bus Financ 70: 102368. https://doi.org/10.1016/j.ribaf.2024.102368 doi: 10.1016/j.ribaf.2024.102368
[40]	Ludvigson SC, Ma S, Ng S (2021) Uncertainty and business cycles: exogenous impulse or endogenous response? Am Econ J Macroecon 13: 369–410. https://doi.org/10.1257/mac.20190171 doi: 10.1257/mac.20190171
[41]	Lundberg SM, Lee SI (2017) A unified approach to interpreting model predictions. Adv Neural Inf Process Syst 30. Available from: https://proceedings.neurips.cc/paper/2017/file/8a20a8621978632d76c43dfd28b67767-Paper.pdf.
[42]	Molnar C (2022) Interpretable Machine Learning: A Guide for Making Black Box Models Explainable. Leanpub, 2nd edition.
[43]	Morikawa M (2016a) Business uncertainty and investment: Evidence from japanese companies. J Macroecon 49: 224–236. https://doi.org/10.1016/j.jmacro.2016.08.001 doi: 10.1016/j.jmacro.2016.08.001
[44]	Morikawa M (2016b) What types of policy uncertainties matter for business? Pac Econ Rev 21: 527–540. https://doi.org/10.1111/1468-0106.12196 doi: 10.1111/1468-0106.12196
[45]	Novy-Marx R (2013) The other side of value: The gross profitability premium. J Financ Econ 108: 1–28. https://doi.org/10.1016/j.jfineco.2013.01.003 doi: 10.1016/j.jfineco.2013.01.003
[46]	Pastor L, Veronesi P (2012) Uncertainty about government policy and stock prices. J Financ 67: 1219–1264. https://doi.org/10.1111/j.1540-6261.2012.01746.x doi: 10.1111/j.1540-6261.2012.01746.x
[47]	Rajgopal S, Venkatachalam M (2011) Financial reporting quality and idiosyncratic return volatility. J Account Econ 51: 1–20. https://doi.org/10.1016/j.jacceco.2010.06.001 doi: 10.1016/j.jacceco.2010.06.001
[48]	Rapach DE, Zhou G (2020) Time‐series and cross‐sectional stock return forecasting: New machine learning methods. Mach Learn Asset Manage: New Dev Financ Appl 1: 1–33. https://doi.org/10.1002/9781119751182.ch1 doi: 10.1002/9781119751182.ch1
[49]	Ren X, Cao Y, Liu PJ, et al. (2023) Does geopolitical risk affect firms' idiosyncratic volatility? evidence from china. Int Rev Financ Anal 90: 102843. https://doi.org/10.1016/j.irfa.2023.102843 doi: 10.1016/j.irfa.2023.102843
[50]	Ribeiro MT, Singh S, Guestrin C (2016) "Why should I trust you?": Explaining the predictions of any classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 1135–1144. https://doi.org/10.1145/2939672.2939778
[51]	Saxegaard ECA, Davis SJ, Ito A, et al. (2022) Policy uncertainty in Japan. J Japanese Int Econ 64: 101192. https://doi.org/10.1016/j.jjie.2022.101192 doi: 10.1016/j.jjie.2022.101192
[52]	Schwert GW (1989) Why does stock market volatility change over time? J Financ 44: 1115–1153. https://doi.org/10.1111/j.1540-6261.1989.tb02647.x doi: 10.1111/j.1540-6261.1989.tb02647.x
[53]	Shin S, Naka A, Wang L (2024) Policy uncertainty and idiosyncratic volatility on nikkei 225 stocks. Appl Econ Lett 32: 1–8. https://doi.org/10.1080/13504851.2024.2311310 doi: 10.1080/13504851.2024.2311310
[54]	Shiratsuka S (2001) Asset prices, financial stability, and monetary policy: based on Japan's experience of the asset price bubble. BIS Pap, 261–84. Available from: https://www.bis.org/publ/bppdf/bispap01l.pdf.
[55]	Siddiqui F, Kong Y, Ali H, et al. (2024) Energy-related uncertainty and idiosyncratic return volatility: Implications for sustainable investment strategies in chinese firms. Sustainability 16: 7423. https://doi.org/10.3390/su16177423 doi: 10.3390/su16177423
[56]	Tabatabaei Poudeh SR, Choi S, Fu C (2022) The effect of covid-19 on the relationship between idiosyncratic volatility and expected stock returns. Risks 10: 57. https://doi.org/10.3390/risks10030057 doi: 10.3390/risks10030057
[57]	Tong C, Huang Z, Wang T, et al. (2023) The effects of economic uncertainty on financial volatility: A comprehensive investigation. J Empir Financ 73: 369–389. https://doi.org/10.1016/j.jempfin.2023.08.004 doi: 10.1016/j.jempfin.2023.08.004
[58]	Xie Z, Ali H, Kumar S, et al. (2024) The impact of energy-related uncertainty on corporate investment decisions in China. Energies 17: 2368. https://doi.org/10.3390/en17102368 doi: 10.3390/en17102368
[59]	Zhang C (2010) A reexamination of the causes of time-varying stock return volatilities. J Financ Quant Anal 45: 663–684. https://doi.org/10.1017/S0022109010000232 doi: 10.1017/S0022109010000232

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)