Melola Discovery

← Melola Discovery

Open access Jurnal

Machine Learning-Based Univariate Time Series Imputation Method for Estimating Missing Values in Non-Stationary Data

Metode Machine Learning-Based Univariate Time Series Imputation Method untuk Estimasi Nilai Hilang pada Data Non-Stasioner

Jurnal Matematika, Statistika dan Komputasi · Vol. 21 No. 1 (2024) · pp. 307–320

DOI
10.20956/j.v21i1.36468
ISSN
2614-8811, 1858-1382
Published
2024-09-15
Publisher
Department of Mathematics, Hasanuddin University
Language
en
License
creativecommons.org/licenses/by/4.0

Abstract

English

Handling missing values in time series data is crucial because they can disrupt data analysis and interpretation. Sequentially missing values in time series often pose a more complex challenge compared to randomly missing values. One of the promising recent methods is Machine Learning-Based Univariate Time Series Imputation (MLBUI), although it is still not widely used and its accessibility is limited. MLBUI employs Random Forest Regression (RFR) and Support Vector Regression (SVR) algorithms. This study evaluates the performance of MLBUI in addressing missing data scenarios in non-stationary univariate time series data. The data used in this research is the average temperature data from Bogor Regency. The missing data scenarios considered include rates of 6%, 10%, and 14%. Besides MLBUI, five other comparison methods are used: Kalman StructTS, Kalman Auto-ARIMA, Spline Interpolation, Stine Interpolation, and Moving Average. The results show that MLBUI performs poorly for non-stationary data, although the obtained Mean Absolute Percentage Error (MAPE) is below 10%.

Bahasa Indonesia

Penanganan nilai hilang dalam data deret waktu adalah krusial karena dapat menyebabkan gangguan dalam analisis dan interpretasi data. Khususnya, nilai-nilai hilang yang terjadi secara berurutan dalam deret waktu seringkali menjadi tantangan yang lebih kompleks dibandingkan dengan nilai-nilai yang hilang secara acak. Salah satu metode terbaru yang menjanjikan adalah Machine Learning-Based Univariate Time Series Imputation (MLBUI), meskipun masih belum banyak digunakan dan aksesibilitasnya masih terbatas. MLBUI menggunakan algoritma adalah Random Forest Regression (RFR) dan Support Vector Regression (SVR). Dalam studi ini, evaluasi dilakukan terhadap kinerja MLBUI dalam mengatasi skenario data yang hilang pada deret waktu univariat yang non-stasioner. Data yang digunakan pada penelitian ini yaitu data suhu rata-rata dari Kabupaten Bogor. Skenario kehilangan data yang dipertimbangkan mencakup tingkat 6%, 10%, dan 14%. Selain MLBUI, lima metode perbandingan lainnya digunakan: Kalman StructTS, Kalman Auto-ARIMA, Interpolasi Spline, Interpolasi Stine, dan Moving Average. Hasil penelitian menunjukkan bahwa MLBUI memberikan hasil yang kurang baik untuk data yang non-stasioner walaupun nilai Mean Absolute Percentage Error (MAPE) yang diperoleh berada dibawah 10%.

Keywords

Collaboration network of the authors

33 authors · 91 co-authorship links · 6 across different organisations

Hover or tap a name to highlight its co-authors. Double-click to open the author's page.

  • IPB University
  • Bandung Islamic University
  • Affiliation unknown
  • Different organisations
  • Same organisation
  • Affiliation unknown
  • Thicker line = more joint articles · bigger circle = more articles
Show as table
Co-authorOrganisationJoint articles
Khairil Anwar NotodiputroBandung Islamic University4
Indahwati IndahwatiAffiliation unknown3
Anwar FitriantoAffiliation unknown2
Anang KurniaAffiliation unknown2
Anik DjuraidahAffiliation unknown2
Herlin FransiskaAffiliation unknown2
Bagus SartonoIPB University1
Leni Anggraini SusantiIPB University1
Teguh PrasetyoIPB University1
Rizki Alifah PutriIPB University1
Yenni AngrainiIPB University1
Andi Harismahyanti A.Affiliation unknown1
Maulida Fajrining TyasAffiliation unknown1
Reni AmeliaAffiliation unknown1
Farah AbqorunnisaAffiliation unknown1
Nindya WulandariAffiliation unknown1
Irzaman IrzamanAffiliation unknown1
Utami Dyah SyafitriAffiliation unknown1
Muhammad Nur AidiAffiliation unknown1
Yesan TiaraAffiliation unknown1
Rika RachmawatiAffiliation unknown1
Nila LestariAffiliation unknown1
Elisa D JuliantiAffiliation unknown1
Indra Rivaldi SiregarAffiliation unknown1
Pardomuan Robinson SihombingAffiliation unknown1
Nadira Nisa AlwaniAffiliation unknown1
Megawati MegawatiAffiliation unknown1
Alfa Nugraha PradanaAffiliation unknown1
Kusman SadikAffiliation unknown1
Uswatun HasanahAffiliation unknown1

Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.

Metadata source: OAI-PMH (metadata jurnal) https://journal.unhas.ac.id/index.php/jmsk/oai, first harvested 5 Nov 2026, last seen 5 Nov 2026.
Original metadata (oai_dc) 
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en">Machine Learning-Based Univariate Time Series Imputation Method for Estimating Missing Values in Non-Stationary Data</dc:title>
	<dc:title xml:lang="id">Metode Machine Learning-Based Univariate Time Series Imputation Method untuk Estimasi Nilai Hilang pada Data Non-Stasioner</dc:title>
	<dc:creator xml:lang="en">Ramadhani, Dini </dc:creator>
	<dc:creator xml:lang="en">Soleh, Agus Mohamad </dc:creator>
	<dc:creator xml:lang="en">Erfiani, Erfiani</dc:creator>
	<dc:subject xml:lang="en">Imputation</dc:subject>
	<dc:subject xml:lang="en">Non-Stationary Data</dc:subject>
	<dc:subject xml:lang="en">Machine Learning</dc:subject>
	<dc:subject xml:lang="en">Missing Values</dc:subject>
	<dc:subject xml:lang="en">data processing</dc:subject>
	<dc:subject xml:lang="en">temperature</dc:subject>
	<dc:subject xml:lang="en">data analysis</dc:subject>
	<dc:subject xml:lang="id">Imputasi</dc:subject>
	<dc:subject xml:lang="id">Data Non-Stasioner</dc:subject>
	<dc:subject xml:lang="id">Pembelajaran Mesin</dc:subject>
	<dc:subject xml:lang="id">Nilai Hilang</dc:subject>
	<dc:subject xml:lang="id">pemrosesan data</dc:subject>
	<dc:subject xml:lang="id">temperatur</dc:subject>
	<dc:subject xml:lang="id">data analisis</dc:subject>
	<dc:description xml:lang="en">Handling missing values in time series data is crucial because they can disrupt data analysis and interpretation. Sequentially missing values in time series often pose a more complex challenge compared to randomly missing values. One of the promising recent methods is Machine Learning-Based Univariate Time Series Imputation (MLBUI), although it is still not widely used and its accessibility is limited. MLBUI employs Random Forest Regression (RFR) and Support Vector Regression (SVR) algorithms. This study evaluates the performance of MLBUI in addressing missing data scenarios in non-stationary univariate time series data. The data used in this research is the average temperature data from Bogor Regency. The missing data scenarios considered include rates of 6%, 10%, and 14%. Besides MLBUI, five other comparison methods are used: Kalman StructTS, Kalman Auto-ARIMA, Spline Interpolation, Stine Interpolation, and Moving Average. The results show that MLBUI performs poorly for non-stationary data, although the obtained Mean Absolute Percentage Error (MAPE) is below 10%.</dc:description>
	<dc:description xml:lang="id">Penanganan nilai hilang dalam data deret waktu adalah krusial karena dapat menyebabkan gangguan dalam analisis dan interpretasi data. Khususnya, nilai-nilai hilang yang terjadi secara berurutan dalam deret waktu seringkali menjadi tantangan yang lebih kompleks dibandingkan dengan nilai-nilai yang hilang secara acak. Salah satu metode terbaru yang menjanjikan adalah Machine Learning-Based Univariate Time Series Imputation (MLBUI), meskipun masih belum banyak digunakan dan aksesibilitasnya masih terbatas. MLBUI menggunakan algoritma adalah Random Forest Regression (RFR) dan Support Vector Regression (SVR). Dalam studi ini, evaluasi dilakukan terhadap kinerja MLBUI dalam mengatasi skenario data yang hilang pada deret waktu univariat yang non-stasioner. Data yang digunakan pada penelitian ini yaitu data suhu rata-rata dari Kabupaten Bogor. Skenario kehilangan data yang dipertimbangkan mencakup tingkat 6%, 10%, dan 14%. Selain MLBUI, lima metode perbandingan lainnya digunakan: Kalman StructTS, Kalman Auto-ARIMA, Interpolasi Spline, Interpolasi Stine, dan Moving Average. Hasil penelitian menunjukkan bahwa MLBUI memberikan hasil yang kurang baik untuk data yang non-stasioner walaupun nilai Mean Absolute Percentage Error (MAPE) yang diperoleh berada dibawah 10%.</dc:description>
	<dc:publisher xml:lang="en">Department of Mathematics, Hasanuddin University</dc:publisher>
	<dc:date>2024-09-15</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en">text</dc:type>
	<dc:type xml:lang="id">teks</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.unhas.ac.id/index.php/jmsk/article/view/36468</dc:identifier>
	<dc:identifier>10.20956/j.v21i1.36468</dc:identifier>
	<dc:source xml:lang="en">Jurnal Matematika, Statistika dan Komputasi; Vol. 21 No. 1 (2024): SEPTEMBER 2024; 307-320</dc:source>
	<dc:source>2614-8811</dc:source>
	<dc:source>1858-1382</dc:source>
	<dc:language>en</dc:language>
	<dc:relation>https://journal.unhas.ac.id/index.php/jmsk/article/view/36468/12016</dc:relation>
	<dc:relation>10.20956/j.v21i1.36468.g12016</dc:relation>
	<dc:coverage xml:lang="en">Bogor</dc:coverage>
	<dc:coverage xml:lang="id">Bogor</dc:coverage>
	<dc:rights xml:lang="en">Copyright (c) 2024 Jurnal Matematika, Statistika dan Komputasi</dc:rights>
	<dc:rights xml:lang="en">https://creativecommons.org/licenses/by/4.0</dc:rights>
</oai_dc:dc>