 |
 |
 |
 |
 |
 |
Sains Malaysiana 54(2)(2025): 437-448
http://doi.org/10.17576/jsm-2025-5402-10
Penanda Aras Integriti Data Kewangan: Penemuan Corak Data Kewangan Indeks FBM KLCI Malaysia Menggunakan Pendekatan Hukum Benford Pareto Diperluas
(Financial Data Integrity Benchmark:
Discovery of Patterns in Financial Data of Malaysia’s FBM KLCI Index Using an
Extended Benford Pareto Law Approach)
SHAR
NIZAM SHARIF, SAIFUL HAFIZAH JAAMAN* & SAIFUL IZZUAN HUSSAIN
Jabatan Sains Matematik, Fakulti Sains dan Teknologi, Universiti Kebangsaan Malaysia,
43600 UKM Bangi, Selangor, Malaysia
Received: 29 May
2024/Accepted: 26 November 2024
Abstrak
Integriti data adalah penting dalam konteks akademik dan praktikal. Hukum Benford dengan prinsip Pareto menilai ketulenan data secara berkesan dengan memodelkan taburan digit pelopor signifikan. Hukum Benford menjangkakan corak taburan logaritma merentas set data yang pelbagai. Penyelidikan ini bertujuan untuk memperluas Hukum Benford Pareto dengan mengoptimumkan parameter bentuk taburan menggunakan kaedah simpleks, meningkatkan kebolehgunaannya sebagai alat forensik untuk mengesan manipulasi data dengan menganalisis penyelewengan daripada taburan digit yang dijangkakan. Kajian memanfaatkan Hukum Benford Pareto Diperluas pada data indeks FBM
KLCI Malaysia menerusi metodologi penyelidikan dwi fasa. Pada mulanya, model dilatih menggunakan data dari 2010 hingga 2020 untuk menentukan taburan jangkaan bagi digit pelopor signifikan. Selepas itu, keberkesanan model diuji dengan set data baharu pada tahun 2020, 2021 dan
2022. Pengesahan model melibatkan ujian keakuran sisihan min mutlak dan khi kuasa dua untuk menilai keakuran kepada prinsip Hukum Benford dan mengesan anomali.
Keputusan mengesahkan bahawa walaupun set data latihan akur kepada Hukum Benford Pareto Diperluas, sisihan min mutlak mengesan sisihan ketara dalam set data uji untuk tahun 2020 dan 2022 mencadangkan potensi manipulasi. Walaupun kajian kes ini memberi tumpuan kepada pasaran saham Malaysia, algoritma yang dibangunkan mempunyai potensi untuk aplikasi yang universal dalam pendekatan analisis forensik data di peringkat global.
Kata kunci: Digit pelopor signifikan; hukum Benford Pareto; pengoptimuman simpleks
Abstract
Data
integrity is crucial in academic and practical contexts. Benford’s Law, rooted in the Pareto principle, effectively assesses data authenticity by modeling the distribution of significant leading digits. Benford’s Law anticipates a logarithmic distribution
pattern across diverse datasets. This study aims to extend the Benford Pareto Law by optimizing distribution shape
parameters using the simplex method, enhancing its applicability as a forensic
tool for detecting data manipulation by analyzing deviations from expected digit distributions. This study applied the Extended Benford Pareto Law to the FBM KLCI Malaysia index data,
employing a dual-phase research methodology. Initially, the model was trained
using data from 2010 to 2020 to determine the expected distribution of
significant leading digits. Subsequently, the model’s effectiveness was tested
with new data sets in 2020, 2021, and 2022. The model evaluation involved
absolute minimum deviation and chi-square tests to assess conformity to Benford’s Law principle and detect anomalies. Results
confirmed that while the training dataset conformed to Extended Pareto Benford’s Law, the minimum absolute deviation test detected
notable deviations in the test datasets for 2020 and 2022 suggesting potential
manipulations. Although this case study focuses on the Malaysian stock market,
the developed algorithm holds global potential for universal application in
data forensic analysis approaches.
Keywords: Pareto Benford’s Law; significant leading digit; simplex optimization
REferences
Buvana, M. & Muthumayil,
K. 2021. Prediction of COVID-19 patient using supervised machine learning
algorithm. Sains Malaysiana 50(8): 2479-2497.
Benford, F. 1938. The law of anomalous numbers. Proceedings
of the American Philosophical Society 78(4): 551-572.
Corazza, M., Ellero, A.
& Zorzi, A. 2010. Checking financial markets via Benford’s law: The S&P 500 case. In Mathematical and
Statistical Methods for Actuarial Sciences and Finance, disunting oleh Corazza, M. & Pizzi,
C. Springer, Milano. pp. 93-103. https://doi.org/10.1007/978-88-470-1481-7
Hairuddin, H.B.M. & Nur-Firyal,
R. 2022. Hubungan antara emas, minyak dan pasaran saham Malaysia semasa perintah kawalan pergerakan COVID-19 menggunakan regresi kuantil dan siri masa. Sains Malaysiana 51(7): 2315-2327.
Hill, T.P. 1995. A statistical derivation
of the sidnificant-digit law. Statististical Science 10(4): 354-363.
Hürlimann, W. 2015. Benford’s law in scientific research. Int. J. Sci. Eng. Res. 6(7):
143-148.
Jayasree, M. 2017. Stock trading and stock returns:
Understanding the distributional properties of the numbers - The evidence from
India Nifty Fifty. Jindal Journal of Business Research 6(2): 171-185.
https://doi.org/10.1177/2278682117727209
Jing, P.K. & Samsudin,
H.B. 2018. Performance of real estate investment trust (REITS) after global
economy crises in Malaysia. Journal of Quality Measurement and Analysis 14(2): 35-44.
Khalil, M.R.A. & Bakar, A.A. 2023. A
comparative study of deep learning algorithms in univariate and multivariate
forecasting of the Malaysian stock market. Sains Malaysiana 52(3): 993-1009.
Kossovsky, A. E. 2014. Benford's law: theory, the general law of relative quantities, and forensic fraud
detection applications (Vol. 3). World Scientific.
Krakar, Z. & Žgela,
M. 2009. Application of Benford’s Law in payment
systems auditing. Journal of Information and Organizational Sciences 33(1): 39-51.
Larsen, J.E. 2017. Benford’s Law and earnings management detection: The case of REITs. Journal of
Forensic & Investigative Accounting 9(2): 779-790.
Ley, E. 1996. On the peculiar distribution
of the U.S. stock indexes’ digits. The American Statistician 50(4): 311.
https://doi.org/10.2307/2684926
Lobão, J. & Pereira, C. 2016. Looking for
psychological barriers in nine European stock market indices. Dutch Journal
of Finance and Management 1(1): 1-14.
https://doi.org/10.20897/lectito.201639
Miller, S. J. (2015). A quick introduction
to Benford’s Law. Benford’s Law: Theory and Applications, 3-22.
Nigrini, M.J. 2012. Benford’s Law: Application for Forensic Accounting, Auditing, and Fraud Dectection. Hoboken, New Jersey: John Wiley & Sons,
Inc.
Nurul Hanis Aminuddin Jafry, Ruzanna Ab Razak & Noriszura Ismail. 2020. Ukuran kebersandaran bagi pulangan lima-minit berbanding pulangan harian menggunakan kopula statik dan dinamik. Sains Malaysiana 49(8): 2023-2034.
Ozaki, Y., Yano, M. & Onishi, M. 2017. Effective hyperparameter optimization
using Nelder-Mead method in deep learning. IPSJ
Transactions on Computer Vision and Applications 9: 1-12.
Rodriguez, R.J. 2004. Reducing false alarms
in the detection of human influence on data. Journal of Accounting, Auditing
& Finance 19(2): 141-158.
Shengmin, Z. & Wenchao,
W. 2010. Does Chinese stock indices agree with Benford’s law? 2010 International Conference on Management and Service Science. pp. 1-3. https://doi.org/10.1109/ICMSS.2010.5575999
Zainol Abidin, S.N., Jaaman, S.H. & Bakar, A.S.A. 2023. The analysis level
of optimism that influence investor’s risk tolerance in asset allocation. Sains Malaysiana 52(10): 2999-3011.
Zulkarnain, N. & Hadi,
M.S.A. 2024. Modelling transmission dynamics of COVID-19 during pre-vaccination
period in Malaysia: A GUI-based seird predictive
model using streamlit. Journal of Quality
Measurement and Analysis 20(1): 25-40.
|
|||
|
