Can we profit from BigTechs' time series models in predicting earnings per share? Evidence from Poland

Wojciech Kuryłek; Wojciech Kuryłek

doi:10.3934/DSFE.2024008

Data Science in Finance and Economics

2024, Volume 4, Issue 2: 218-235. doi: 10.3934/DSFE.2024008

Previous Article Next Article

Research article

Can we profit from BigTechs' time series models in predicting earnings per share? Evidence from Poland

Wojciech Kuryłek ^,

Faculty of Management, University of Warsaw, 1/3 Szturmowa Street, 02-678 Warsaw, Poland

Received: 03 January 2024 Revised: 01 March 2024 Accepted: 20 March 2024 Published: 15 April 2024
JEL Codes: C01, C02, C12, C14, C58, G17

Forecasting earnings for publicly traded companies is of paramount significance for investments, which is the background of this research. This holds particularly true in emerging markets where the coverage of these companies by financial analysts' predictions is limited. This research investigation delves into the prediction inaccuracies of cutting-edge time series forecasting algorithms created by major technology companies such as Facebook, LinkedIn, Amazon, and Google. These techniques are employed to analyze earnings per share data for publicly traded Polish companies during the period spanning from the financial crisis to the pandemic shock. My objective was to compare prediction errors of analyzed models, using scientifically defined error measures and a series of statistical tests. The seasonal random walk model demonstrated the lowest error of prediction, which might be attributable to the overfitting of complex models.
- earnings per share,
- financial forecasting,
- machine learning,
- deep learning,
- Prophet by Facebook,
- SilverKite by LinkedIn,
- DeepAR by Amazon,
- TFT by Google,
- Warsaw Stock Exchange
Citation: Wojciech Kuryłek. Can we profit from BigTechs' time series models in predicting earnings per share? Evidence from Poland[J]. Data Science in Finance and Economics, 2024, 4(2): 218-235. doi: 10.3934/DSFE.2024008

Related Papers:

Abstract

Forecasting earnings for publicly traded companies is of paramount significance for investments, which is the background of this research. This holds particularly true in emerging markets where the coverage of these companies by financial analysts' predictions is limited. This research investigation delves into the prediction inaccuracies of cutting-edge time series forecasting algorithms created by major technology companies such as Facebook, LinkedIn, Amazon, and Google. These techniques are employed to analyze earnings per share data for publicly traded Polish companies during the period spanning from the financial crisis to the pandemic shock. My objective was to compare prediction errors of analyzed models, using scientifically defined error measures and a series of statistical tests. The seasonal random walk model demonstrated the lowest error of prediction, which might be attributable to the overfitting of complex models.

References

[1]	Ahmadpour A, Etemadi H, Moshashaei S, et al. (2015) Earnings Per Share Forecast Using Extracted Rules from Trained Neural Network by Genetic Algorithm. Comput Econ 46: 55–63. https://doi.org/10.1007/s10614-014-9455-6 doi: 10.1007/s10614-014-9455-6
[2]	Aidan IA, Al-Jeznawi D, Al-Zwainy FM, et al. (2020) Predicting earned value indexes in residential complexes' construction projects using artificial neural network model. Int J Intell Eng Syst 13: 248–259. http://dx.doi.org/10.22266/ijies2020.0831.22 doi: 10.22266/ijies2020.0831.22
[3]	Alexander RA, Govern DM (1994) A New and Simpler Approximation for ANOVA under Variance Heterogeneity. J Educ Stat 19: 91–101. http://dx.doi.org/10.2307/1165140 doi: 10.2307/1165140
[4]	Ahammad P, Al Orjany SE, Chen A, et al. (2022) Greykite: deploying flexible forecasting at scale at LinkedIn. Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 3007–3017. https://doi.org/10.1145/3534678.3539165
[5]	Al-Somaydaii JA, Al-Zwainy FM, Hammoody O, et al. (2022) Forecasting and determining of cost performance index of tunnels projects using artificial neural networks. Int J Comput Civil Struct Eng 18: 51–60. https://doi.org/10.22337/2587-9618-2022-18-1-51-60 doi: 10.22337/2587-9618-2022-18-1-51-60
[6]	Al-Zwainy FM, Amer R, Khaleel T, et al. (2016) Reviewing of the simulation models in cost management of the construction projects. Civil Eng J 2: 607–622. https://doi.org/10.28991/cej-2016-00000063 doi: 10.28991/cej-2016-00000063
[7]	AL-Zwainy FM, Huda F, Ibrahim A, et al. (2018) Development of an Analytical Software for Cost Estimation for Highway Project. Int J Appl Eng Res 13: 6944–6951.
[8]	Al-Zwainy FM, Jaber FK, Jasim NA, et al. (2020) Forecasting techniques in construction industry: earned value indicators and performance models. Sci Rev Eng Environ Sci 29: 234–243. https://doi.org/10.22630/PNIKS.2020.29.2.20 doi: 10.22630/PNIKS.2020.29.2.20
[9]	Al-Zwainy FM, Raheem SH (2020) Innovation of Analytical Software for Financing Construction Projects: Infrastructure Projects in Iraq as a Case Study, In: IOP Conference Series: Materials Science and Engineering, IOP Publishing, 978: 012015. https://doi.org/10.1088/1757-899X/978/1/012015
[10]	Arık SÖ, Lim B, Loeff N, et al. (2021) Temporal Fusion Transformers for interpretable multi-horizon time series forecasting. Int J Forecast 37: 1748–1764. https://doi.org/10.1016/j.ijforecast.2021.03.012 doi: 10.1016/j.ijforecast.2021.03.012
[11]	Ball R, Watts R (1972) Some Time Series Properties of Accounting Income. J Financ 27: 663–681. http://dx.doi.org/10.1111/j.1540-6261.1972.tb00991.x doi: 10.1111/j.1540-6261.1972.tb00991.x
[12]	Bathke Jr AW, Lorek KS (1984) The Relationship between Time-Series Models and the Security Market's Expectation of Quarterly Earnings. Account Rev 59: 163–176.
[13]	Bengio Y, Courville A, Goodfellow I, et al. (2017) Deep Learning. Cambridge, Massachusetts: The MIT Press.
[14]	Bradshaw M, Drake M, Myers J, et al. (2012) A re-examination of analysts' superiority over time-series forecasts of annual earnings. Rev Account Stud 17: 944–968. http://dx.doi.org/10.1007/s11142-012-9185-8 doi: 10.1007/s11142-012-9185-8
[15]	Brandon Ch, Jarrett JE, Khumawala SB, et al. (1983) On the predictability of corporate earnings per share. J Bus Financ Account 10: 373–387.
[16]	Brandon Ch, Jarrett JE, Khumawala SB, et al. (1986) Comparing forecast accuracy for exponential smoothing models of earnings-per-share data for financial decision making. Decision Sci 17: 186–194. http://dx.doi.org/10.1111/j.1540-5915.1986.tb00220.x doi: 10.1111/j.1540-5915.1986.tb00220.x
[17]	Brandon Ch, Jarrett JE, Khumawala SB, et al. (1987) A Comparative Study of the Forecasting Accuracy of Holt‐Winters and Economic Indicator Models of Earnings Per Share For Financial Decision Making. Manag Financ 13: 10–15. http://dx.doi.org/10.1108/eb013581 doi: 10.1108/eb013581
[18]	Brooks LD, Buckmaster DA (1976) Further Evidence of The Time Series Properties Of Accounting Income. J Financ 31: 1359–1373. http://dx.doi.org/10.1111/j.1540-6261.1976.tb03218.x doi: 10.1111/j.1540-6261.1976.tb03218.x
[19]	Brown LD, Griffin PA, Hagerman RL, et al. (1987) Security analyst superiority relative to univariate time-series models in forecasting quarterly earnings. J Account Econ 9: 61–87. http://dx.doi.org/10.1016/0165-4101(87)90017-6 doi: 10.1016/0165-4101(87)90017-6
[20]	Brown LD, Rozeff MS (1979) Univariate Time-Series Models of Quarterly Accounting Earnings per Share: A Proposed Model. J Account Res 17: 179–189. http://dx.doi.org/10.2307/2490312 doi: 10.2307/2490312
[21]	Cao Q, Gan Q (2009) Forecasting EPS of Chinese listed companies using a neural network with genetic algorithm. 15th Americas Conference on Information Systems 2009, AMCIS 2009, 2791–2981.
[22]	Cao Q, Parry M (2009) Neural network earnings per share forecasting models: A comparison of backward propagation and the genetic algorithm. Decis Support Syst 47: 32–41. http://dx.doi.org/10.1016/j.dss.2008.12.011 doi: 10.1016/j.dss.2008.12.011
[23]	Cao Q, Schniederjans MJ, Zhang W, et al. (2004) Neural network earnings per share forecasting models: A comparative analysis of alternative methods. Decision Sci 35: 205–237. https://doi.org/10.1111/j.00117315.2004.02674.x doi: 10.1111/j.00117315.2004.02674.x
[24]	Conroy R, Harris T (1987) Consensus Forecasts of Corporate Earnings: Analysts' Forecasts and Time Series Methods. Manag Sci 33: 725–738. http://dx.doi.org/10.1287/mnsc.33.6.725 doi: 10.1287/mnsc.33.6.725
[25]	Corder GW, Foreman DI (2009) Comparing More than Two Unrelated Samples: The Kruskal–Wallis H-Test. In: Nonparametric Statistics for Non-Statisticians, John Wiley & Sons, Hoboken, New Jersey, 99–121. http://dx.doi.org/10.1002/9781118165881
[26]	Dreher S, Eichfelder S, Noth F, et al. (2024) Does IFRS information on tax loss carryforwards and negative performance improve predictions of earnings and cash flows? J Bus Econ 94: 1–39. http://dx.doi.org/10.1007/s11573-023-01147-7 doi: 10.1007/s11573-023-01147-7
[27]	Elend L, Kramer O, Lopatta J, et al. (2020) Earnings prediction with deep learning. German Conference on Artificial Intelligence (Künstliche Intelligenz), KI 2020: Advances in Artificial Intelligence, 267–274. http://dx.doi.org/10.1007/978-3-030-58285-2_22
[28]	Elton EJ, Gruber MJ (1972) Earnings Estimates and the Accuracy of Expectational Data. Manag Sci 18: B409–B424. http://dx.doi.org/10.1287/mnsc.18.8.B409 doi: 10.1287/mnsc.18.8.B409
[29]	Finger CA (1994) The ability of earnings to predict future earnings and cash flow. J Account Res 32: 210–223. http://dx.doi.org/10.2307/2491282 doi: 10.2307/2491282
[30]	Flagmeier V (2022) The information content of deferred taxes under IFRS. Eu Account Rev 31: 495–518. http://dx.doi.org/10.1080/09638180.2020.1826338 doi: 10.1080/09638180.2020.1826338
[31]	Foster G (1977) Quarterly Accounting Data: Time-Series Properties and Predictive-Ability Results. Account Rev 52: 1–21.
[32]	Flunkert V, Gasthaus J, Januschowski T, et al. (2020) DeepAR: Probabilistic forecasting with autoregressive recurrent networks. Int J Forecast 36: 1181–1191. https://doi.org/10.1016/j.ijforecast.2019.07.001 doi: 10.1016/j.ijforecast.2019.07.001
[33]	Gaio L, Gatsios R, Lima F, et al. (2021) Re-examining analyst superiority in forecasting results of publicly-traded Brazilian companies. Revista de Administracao Mackenzie 22: eRAMF210164. https://doi.org/10.1590/1678-6971/eramf210164 doi: 10.1590/1678-6971/eramf210164
[34]	Gerakos J, Gramacy RB (2013) Regression-Based Earnings Forecasts. Chicago Booth Research Paper, 12–26. https://doi.org/10.2139/ssrn.2112137 doi: 10.2139/ssrn.2112137
[35]	Griffin P (1977) The Time-Series Behavior of Quarterly Earnings: Preliminary Evidence. J Account Res 15: 71–83. http://dx.doi.org/10.2307/2490556 doi: 10.2307/2490556
[36]	Grigaliūnienė Ž (2013) Time-series models forecasting performance in the Baltic stock market. Organ Market Emerg E 4: 104–120.
[37]	Holt Ch C (1957) Forecasting seasonals and trends by exponentially weighted moving averages. Working Paper, Carnegie Institute of Technology.
[38]	Holt Ch C (2004) Forecasting seasonals and trends by exponentially weighted moving averages. J Econ Soc Meas 29: 123–125. https://doi.org/10.1016/j.ijforecast.2003.09.015 doi: 10.1016/j.ijforecast.2003.09.015
[39]	Jarrett JE (2008) Evaluating Methods for Forecasting Earnings Per Share. Manag Financ 16: 30–35. http://dx.doi.org/10.1108/eb013647 doi: 10.1108/eb013647
[40]	Johnson TE, Schmitt TG (1974) Effectiveness of Earnings Per Share Forecasts. Financ Manag 3: 64–72. http://dx.doi.org/10.2307/3665292 doi: 10.2307/3665292
[41]	Kim S, Kim H (2016) A new metric of absolute percentage error for intermittent demand forecasts. Int J Forecast 32: 669–679. http://dx.doi.org/10.1016/j.ijforecast.2015.12.003 doi: 10.1016/j.ijforecast.2015.12.003
[42]	Kuryłek W (2023a) The modeling of earnings per share of Polish companies for the post-financial crisis period using random walk and ARIMA models. J Bank Financ Econ 1: 26–43. http://dx.doi.org/10.7172/2353-6845.jbfe.2023.1.2 doi: 10.7172/2353-6845.jbfe.2023.1.2
[43]	Kuryłek W (2023b) Can exponential smoothing do better than seasonal random walk for earnings per share forecasting in Poland? Bank Credit 54: 651–672.
[44]	Lacina M, Lee B, Xu R, et al. (2011) An evaluation of financial analysts and naï ve methods in forecasting long-term earnings, In: K. D Lawrence & R. K. Klimberg (Eds.), Advances in business and management forecasting, Bingley, UK: Emerald, 77–101. http://dx.doi.org/10.1108/S1477-4070(2011)0000008009
[45]	Lai S, Li H (2006) The predictive power of quarterly earnings per share based on time series and artificial intelligence model. Appl Financ Econ 16: 1375–1388. http://dx.doi.org/10.1080/09603100600592752 doi: 10.1080/09603100600592752
[46]	Letham B, Taylor SJ (2018) Forecasting at scale. Am Stat 72: 37–45. https://doi.org/10.7287/peerj.preprints.3190v1 doi: 10.7287/peerj.preprints.3190v1
[47]	Lev B, Souginannis T (2010) The usefulness of accounting estimates for predicting cash flows and earnings. Rev Account Stud 15: 779–807. http://dx.doi.org/10.1007/s11142-009-9107-6 doi: 10.1007/s11142-009-9107-6
[48]	Lorek KS (1979) Predicting Annual Net Earnings with Quarterly Earnings Time-Series Models. J Account Res 17: 190–204. http://dx.doi.org/10.2307/2490313 doi: 10.2307/2490313
[49]	Lorek KS, Willinger GL (1996) A multivariate time-series model for cash-flow data. Account Rev 71: 81–101.
[50]	Lorek KS, Willinger GL (2007) The contextual nature of the predictive power of statistically-based quarterly earnings models. Rev Q Financ Account 28: 1–22. http://dx.doi.org/10.1007/s11156-006-0001-z doi: 10.1007/s11156-006-0001-z
[51]	Lowry R (2014) Concepts and Applications of Inferential Statistics, Chapter 14. Available from: https://onlinebooks.lib, rary.upenn.edu/webbin/book/lookupid?key = olbp66608.
[52]	Pagach DP, Warr RS (2020) Analysts versus time-series forecasts of quarterly earnings: A maintained hypothesis revisited. Adv Account 51: 1–15. http://dx.doi.org/10.1016/j.adiac.2020.100497 doi: 10.1016/j.adiac.2020.100497
[53]	Ruland W (1980) On the Choice of Simple Extrapolative Model Forecasts of Annual Earnings. Financ Manag 9: 30–37. http://dx.doi.org/10.2307/3665165 doi: 10.2307/3665165
[54]	Watts RL (1975) The Time Series Behavior of Quarterly Earnings. Working Paper, Department of Commerce, University of New Castle.
[55]	Wilcoxon F (1945) Individual comparisons by ranking methods. Biometrics Bulletin 1: 80–83. http://dx.doi.org/10.2307/3001968 doi: 10.2307/3001968
[56]	Xiaoqiang W (2022) Research on enterprise financial performance evaluation method based on data mining. In: 2022 IEEE 2nd International Conference on Electronic Technology, Communication and Information (ICETCI). https://doi.org/10.1109/icetci55101.2022.9832404

Reader Comments

Your name:*

Email:*
© 2024 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)