 |
 |
 |
 |
 |
 |
Sains Malaysiana 54(6)(2025): 1617-1628
http://doi.org/10.17576/jsm-2025-5406-16
Detection
Procedure of Structural Changes in State-Space Models: Impulse and Steps
Indicator Saturation Technique
(Prosedur Pengesanan Perubahan Struktur dalam Model Ruang-Keadaan: Teknik Ketepuan Penunjuk Impuls dan Langkah)
FARID ZAMANI CHE ROSE1,*, MOHD
TAHIR ISMAIL2, MUHAMMAD ASLAM SAFARI1, NUR AQILAH
KHADIJAH ROSILI3 & MUHAMMAD FADHIL MARSANI2
1Department of Mathematics and Statistics, Faculty of Science, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor,
Malaysia
2School of Mathematical Sciences, Universiti Sains Malaysia, USM Minden, 11800 Pulau Pinang, Malaysia
3Faculty of Computing and Engineering, Quest International
University, 30250 Ipoh, Perak, Malaysia
Received: 18
July 2024/Accepted: 13 March 2025
Abstract
The presence of extreme structural change in a
sequence of data points over time may have a detrimental impact on the
estimation of economic and financial indicators. Anomalies caused by these
extreme values can distort the estimated parameters, diminish the accuracy of
the time series model, and potentially lead to inaccurate forecasts. In this
research, a general-to-specific modeling approach is
utilized to identify the structural changes through indicator saturation within
the framework of a state-space models as an alternative to current method. By
focusing on impulse and steps indicator saturation, this study evaluates their
effectiveness through Monte Carlo simulations that are replicated 1000 times.
The Monte Carlo experiments demonstrate that the efficiency of indicator
saturation is heavily dependent on factors such as the magnitude of the structural
change, the level of statistical significance, and the position of an extreme
value within the series. Subsequently, this study employs the combined impulse
and steps indicator saturation to detect structural breaks in the FTSE 100 daily
closed stock price index. The most important findings relate to the coefficients for the
structural breaks at
Keywords: General-to-specific; indicator saturation; model
selection; Monte Carlo; structural changes
Abstrak
Kehadiran perubahan struktur yang ketara dalam satu data siri
masa boleh memberi kesan buruk terhadap penganggaran penunjuk ekonomi dan
kewangan. Nilai melampau yang menyebabkan anomali boleh memesongkan parameter
yang dianggarkan, mengurangkan ketepatan model siri masa dan berpotensi
menghasilkan ramalan yang tidak tepat. Dalam kajian ini, pendekatan pemodelan
umum kepada khusus digunakan untuk mengenal pasti perubahan struktur melalui
petunjuk ketepuan dalam kerangka model ruang keadaan sebagai alternatif kepada
kaedah semasa. Dengan memberi tumpuan kepada petunjuk ketepuan impuls dan
langkah, kajian ini menilai keberkesanannya melalui simulasi Monte Carlo yang
diulang sebanyak 1000 kali. Uji kaji Monte Carlo menunjukkan bahawa kecekapan
ketepuan penunjuk sangat bergantung kepada faktor seperti magnitud perubahan
struktur, tahap signifikan statistik dan kedudukan nilai ekstrem dalam siri
data tersebut. Seterusnya, kajian ini menggunakan gabungan ketepuan penunjuk
impuls dan langkah untuk mengesan perubahan struktur dalam indeks harga penutup
harian FTSE 100. Penemuan paling penting berkaitan dengan pekali bagi perubahan
struktur pada penemuan ini menunjukkan bahawa ciri, kedudukan dan arah nilai
melampau yang dikesan oleh ketepuan penunjuk impuls bertepatan dengan berlakunya
pandemik COVID-19, yang telah memberi kesan global terhadap aktiviti ekonomi.
Penemuan ini boleh membawa kepada pemahaman yang lebih baik tentang bagaimana
pasaran saham di UK bertindak balas terhadap dasar kerajaan akibat pandemik
COVID-19.
Kata kunci: Monte Carlo; pemilihan model; perubahan
struktur; petunjuk ketepuan; umum kepada tertentu
REFERENCES
Alimohammadi, H. & Nancy Chen, S. 2022. Performance
evaluation of outlier detection techniques in production timeseries: A
systematic review and meta-analysis. Expert Systems with Applications 191: 116371. https://doi.org/10.1016/J.ESWA.2021.116371
Bakar, N.A.
2019. Monte Carlo simulation for data volatility analysis of stock prices in Islamic
finance for Malaysia composite index. International Journal of Advanced
Engineering Research and Science 6(3): 6-12. https://doi.org/10.22161/ijaers.6.3.2
Bergman, L.
& Hoshen, Y. 2020. Classification-Based
Anomaly Detection for General Data. http://arxiv.org/abs/2005.02359
Castle, J.L.
& Kurita, T. 2021. A dynamic econometric analysis of the dollar-pound
exchange rate in an era of structural breaks and policy regime shifts. Journal
of Economic Dynamics and Control 128: 104139. https://doi.org/10.1016/J.JEDC.2021.104139
Castle, J.L.
& Hendry, D.F. 2019. Modelling our Changing World. Palgrave
Macmillan. https://doi.org/10.1007/978-3-030-21432-6
Castle, J.L.,
Hendry, D.F. & Martinez, A.B. 2023. The historical role of energy in UK
inflation and productivity with implications for price inflation. Energy
Economics 126: 106947. https://doi.org/10.1016/j.eneco.2023.106947
Castle, J.L.,
Doornik, J.A. & Hendry, D.F. 2012. Model selection when there are multiple
breaks. Journal of Econometrics 169(2): 239-246. https://doi.org/10.1016/j.jeconom.2012.01.026
Castle, J.L.,
Doornik, J.A. & Hendry, D.F. 2011. Evaluating automatic model selection. Journal
of Time Series Econometrics 3(1): 8-8. https://doi.org/10.2202/1941-1928.1097
Castle, J.L.,
Doornik, J.A., Hendry, D.F. & Pretis, F. 2015.
Detecting location shifts by step-indicator saturation. Econometrics 3(2): 240-264. https://doi.org/10.3390/econometrics03020240
Che Rose, F.Z.,
Ismail, M.T. & Tumin, M.H. 2021. Outliers detection in state-space model
using indicator saturation approach. Indonesian Journal of Electrical
Engineering and Computer Science 22(3): 1688-1696. https://doi.org/10.11591/ijeecs.v22.i3.pp1688-1696
Doornik, J.A.,
Hendry, D.F. & Pretis, F. 2013. Step-Indicator
Saturation. Oxford University, Department of Economics. Discussion
Paper Series.
Durbin, J. & Koopman, S.J. 2012. Time
Series Analysis by State Space Methods. 2nd ed. Oxford:
Oxford University Press.
Harvey, A.
& Koopman, S.J. 2000. Signal extraction and the formulation of unobserved
components models. The Econometrics Journal 3(1): 84-107. https://doi.org/10.1111/1368-423x.00040
Hendry, D.F.
1999. An econometric analysis of US food expenditure. In Methodology and Tacit
Knowledge: Two Experiments in Econometrics, edited by Magnus, J.R. & Morgab, M.S. New York: John Wiley & Sons.
Huang, J., Cheng, D. & Zhang, S. 2023. A
novel outlier detecting algorithm based on the outlier turning points. Expert
Systems with Applications 231: 120799. https://doi.org/10.1016/j.eswa.2023.120799
Johansen, S.
& Nielsen, B. 2009. An analysis of the indicator saturation estimator as a
robust regression estimator. The Methodology and Practice of Econometrics: A
Festschrift in Honour of David F. Hendry, edited by Castle, J. &
Shephard, N. Oxford. https://doi.org/10.1093/acprof:oso/9780199237197.003.0001
Marczak, M.
& Proietti, T. 2016. Outlier detection in structural time series models:
The indicator saturation approach. International Journal of Forecasting 32(1): 180-202. https://doi.org/10.1016/j.ijforecast.2015.04.005
Mariscal, R.
& Powell, A. 2014. Commodity price booms and breaks: Detection, magnitude
and implications for developing countries. IDB Working Paper No. IDB-WP-444. SSRN. https://doi.org/10.2139/ssrn.2384422
Pellini, E.
2021. Estimating income and price elasticities of residential electricity
demand with autometrics. Energy Economics 101:
105411. https://doi.org/10.1016/J.ENECO.2021.105411
Pretis, F., Reade, J.J. & Sucarrat,
G. 2018. Automated general-to-specific (GETS) regression modeling and indicator saturation for outliers and structural breaks. Journal of
Statistical Software 86(3): 1-44. https://doi.org/10.18637/jss.v086.i03
Pretis, F., Schneider, L., Smerdon, J.E. & Hendry, D.F. 2016. Detecting volcanic eruptions in temperature
reconstructions by designed break-indicator saturation. Journal of Economic
Surveys 30(3): 403-429. https://doi.org/10.1111/joes.12148
Santos, C., Hendry, D.F. & Johansen, S. 2008. Automatic selection of indicators in a fully
saturated regression. Computational Statistics 23(2): 317-335. https://doi.org/10.1007/s00180-007-0054-z
Sewwandi, M.A.N.D., Li, Y. & Zhang, J. 2024.
k-outlier removal based on contextual label information and cluster purity for
continuous data classification. Expert Systems with Applications 237(Part
C): 121347. https://doi.org/10.1016/j.eswa.2023.121347
Shao, C.,
Zheng, S., Gu, C., Hu, Y. & Qin, X. 2022. A novel outlier detection method
for monitoring data in dam engineering. Expert Systems with Applications 193: 116476. https://doi.org/10.1016/j.eswa.2021.116476
Uzabaci, E., Ercan, I. & Alpu,
O. 2020. Evaluation of outlier detection method performance in symmetric
multivariate distributions. Communications in Statistics: Simulation and
Computation 49(2): 516-531. https://doi.org/10.1080/03610918.2018.1487068
Vishwakarma,
G.K., Paul, C. & Elsawah, A.M. 2020. An algorithm
for outlier detection in a time series model using backpropagation neural
network. Journal of King Saud University - Science 32(8): 3328-3336. https://doi.org/10.1016/j.jksus.2020.09.018
Yu, Y., Zhu, Y., Li, S. & Wan, D. 2014. Mathematical Problems in Engineering 2014: 879736. https://doi.org/10.1155/2014/879736
*Corresponding author; email: faridzamani@upm.edu.my
|
|||
|
