<?xml version="1.0" encoding="ISO-8859-1"?><article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<front>
<journal-meta>
<journal-id>0379-3982</journal-id>
<journal-title><![CDATA[Revista Tecnología en Marcha]]></journal-title>
<abbrev-journal-title><![CDATA[Tecnología en Marcha]]></abbrev-journal-title>
<issn>0379-3982</issn>
<publisher>
<publisher-name><![CDATA[Instituto Tecnológico de Costa Rica]]></publisher-name>
</publisher>
</journal-meta>
<article-meta>
<article-id>S0379-39822024000300033</article-id>
<article-id pub-id-type="doi">10.18845/tm.v37i2.6746</article-id>
<title-group>
<article-title xml:lang="es"><![CDATA[Comparative analysis of traditional methods and a deep learning approach for multivariate imputation of missing values in the meteorological field]]></article-title>
<article-title xml:lang=""><![CDATA[Análisis comparativo de algoritmos tradicionales y un modelo de aprendizaje profundo para la imputación multivariada de valores faltantes en el campo meteorológico]]></article-title>
</title-group>
<contrib-group>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Arias-Muñoz]]></surname>
<given-names><![CDATA[Ana Cristina]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
</contrib>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Cob-García]]></surname>
<given-names><![CDATA[Susana]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
</contrib>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Calvo-Valverde]]></surname>
<given-names><![CDATA[Luis-Alexander]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
</contrib>
</contrib-group>
<aff id="Af1">
<institution><![CDATA[,Instituto Tecnológico de Costa Rica  ]]></institution>
<addr-line><![CDATA[ ]]></addr-line>
<country>Costa Rica</country>
</aff>
<aff id="Af2">
<institution><![CDATA[,Instituto Tecnológico de Costa Rica  ]]></institution>
<addr-line><![CDATA[ ]]></addr-line>
<country>Costa Rica</country>
</aff>
<aff id="Af3">
<institution><![CDATA[,Instituto Tecnológico de Costa Rica  ]]></institution>
<addr-line><![CDATA[ ]]></addr-line>
<country>Costa Rica</country>
</aff>
<pub-date pub-type="pub">
<day>00</day>
<month>09</month>
<year>2024</year>
</pub-date>
<pub-date pub-type="epub">
<day>00</day>
<month>09</month>
<year>2024</year>
</pub-date>
<volume>37</volume>
<numero>3</numero>
<fpage>33</fpage>
<lpage>47</lpage>
<copyright-statement/>
<copyright-year/>
<self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_arttext&amp;pid=S0379-39822024000300033&amp;lng=en&amp;nrm=iso"></self-uri><self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_abstract&amp;pid=S0379-39822024000300033&amp;lng=en&amp;nrm=iso"></self-uri><self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_pdf&amp;pid=S0379-39822024000300033&amp;lng=en&amp;nrm=iso"></self-uri><abstract abstract-type="short" xml:lang="es"><p><![CDATA[Resumen Las observaciones climáticas son la base para varias aplicaciones del mundo real, como el pronóstico del tiempo, el monitoreo del cambio climático y las evaluaciones de impacto ambiental. Sin embargo, la mayoría de los datos son medidos y registrados por dispositivos externos expuestos a numerosas variables, causantes de mal funcionamiento de los dispositivos y, por lo tanto, de los valores faltantes. En la actualidad, se ha investigado en profundidad la imputación de datos en el campo de las series temporales y se han propuesto una gran variedad de métodos, donde predominan los algoritmos tradicionales de clasificación y regresión, no obstante, también existen enfoques de aprendizaje profundo que logran capturar relaciones temporales entre observaciones. En este artículo se realiza un análisis comparativo entre un algoritmo de clasificación, un algoritmo de regresión y un modelo de aprendizaje profundo: algoritmo MissForest, basado en árboles aleatorios; Expectation Maximization with Bootstrap (EMB), el algoritmo de estimación de máxima verosimilitud; y una propuesta de un modelo de aprendizaje profundo, basado en la arquitectura Long-Short Term Memory (LSTM). Se utilizaron datos del campo meteorológico de Costa Rica, los cuales consisten en datos multivariados provenientes de varias estaciones meteorológicas en una misma zona geográfica.]]></p></abstract>
<abstract abstract-type="short" xml:lang="en"><p><![CDATA[Abstract Climate observations are the groundwork for several real-world applications such as weather forecasting, climate change monitoring and environmental impact assessments. However, the data is mostly measured and recorded by external devices exposed to numerous variables, causatives of malfunctions and, therefore, missing values. Nowadays, data imputation in the time series field has been researched in depth and a wide variety of methods have been proposed, where traditional classification and regression algorithms predominate, even though there are also deep learning approaches that manage to capture temporal relationships between observations. In this article, a comparative analysis between a classification imputation algorithm, a regression imputation algorithm, and a deep learning imputation model is made: MissForest algorithm, based on random trees; Expectation Maximization with Bootstrap (EMB), the maximum likelihood estimation algorithm; and a proposed deep learning model, based on the Long-Short Term Memory (LSTM) architecture. Data from the Costa Rica meteorological field were used, which consist of multivariate data coming from several weather stations in the same geographical area.]]></p></abstract>
<kwd-group>
<kwd lng="en"><![CDATA[Data imputation]]></kwd>
<kwd lng="en"><![CDATA[EMB]]></kwd>
<kwd lng="en"><![CDATA[MissForest]]></kwd>
<kwd lng="en"><![CDATA[LSTM]]></kwd>
<kwd lng="en"><![CDATA[time series]]></kwd>
<kwd lng="es"><![CDATA[Imputación de datos]]></kwd>
<kwd lng="es"><![CDATA[EMB]]></kwd>
<kwd lng="es"><![CDATA[MissForest]]></kwd>
<kwd lng="es"><![CDATA[LSTM]]></kwd>
<kwd lng="es"><![CDATA[series de tiempo]]></kwd>
</kwd-group>
</article-meta>
</front><back>
<ref-list>
<ref id="B1">
<label>(1)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Zhang]]></surname>
<given-names><![CDATA[Y.]]></given-names>
</name>
<name>
<surname><![CDATA[Thorburn]]></surname>
<given-names><![CDATA[P. J.]]></given-names>
</name>
<name>
<surname><![CDATA[Xiang and P. Fitch]]></surname>
<given-names><![CDATA[W.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[SSIM&#8212;A Deep Learning Approach for Recovering Missing Time Series Sensor Data]]></article-title>
<source><![CDATA[IEEE Internet of Things Journal]]></source>
<year>2019</year>
</nlm-citation>
</ref>
<ref id="B2">
<label>(2)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Bokde]]></surname>
<given-names><![CDATA[N.]]></given-names>
</name>
<name>
<surname><![CDATA[Beck]]></surname>
<given-names><![CDATA[M. W.]]></given-names>
</name>
<name>
<surname><![CDATA[Martínez-Alvarez and K. Kulat]]></surname>
<given-names><![CDATA[F.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[A novel imputation methodology for time series based on pattern sequence forecasting]]></article-title>
<source><![CDATA[Pattern Recognition Letters]]></source>
<year>2018</year>
</nlm-citation>
</ref>
<ref id="B3">
<label>(3)</label><nlm-citation citation-type="journal">
<article-title xml:lang=""><![CDATA[A Guide to Recurrent Neural Networks: Understanding RNN and LSTM Networks]]></article-title>
<source><![CDATA[builtin.com/data-science/recurrent-neural-networks-and-lstm]]></source>
<year>2022</year>
</nlm-citation>
</ref>
<ref id="B4">
<label>(4)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Jerez]]></surname>
<given-names><![CDATA[J. M.]]></given-names>
</name>
<name>
<surname><![CDATA[Molina]]></surname>
<given-names><![CDATA[I.]]></given-names>
</name>
<name>
<surname><![CDATA[García-Laencina]]></surname>
<given-names><![CDATA[P. J.]]></given-names>
</name>
<name>
<surname><![CDATA[Alba]]></surname>
<given-names><![CDATA[E.]]></given-names>
</name>
<name>
<surname><![CDATA[Ribelles]]></surname>
<given-names><![CDATA[N.]]></given-names>
</name>
<name>
<surname><![CDATA[Martín and L. Franco]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Missing data imputation using statistical and machine learning methods in a real breast cancer problem]]></article-title>
<source><![CDATA[Artificial Intelligence in Medicine]]></source>
<year>2010</year>
</nlm-citation>
</ref>
<ref id="B5">
<label>(5)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Liu]]></surname>
<given-names><![CDATA[T.]]></given-names>
</name>
<name>
<surname><![CDATA[Wei]]></surname>
<given-names><![CDATA[H.]]></given-names>
</name>
<name>
<surname><![CDATA[Zhang]]></surname>
<given-names><![CDATA[K.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Wind power prediction with missing data using Gaussian process regression and multiple imputation]]></article-title>
<source><![CDATA[Applied Soft Computing]]></source>
<year>2018</year>
</nlm-citation>
</ref>
<ref id="B6">
<label>(6)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Quinteros]]></surname>
<given-names><![CDATA[M. E.]]></given-names>
</name>
<name>
<surname><![CDATA[Blazquez]]></surname>
<given-names><![CDATA[C.]]></given-names>
</name>
<name>
<surname><![CDATA[Ossa]]></surname>
<given-names><![CDATA[X.]]></given-names>
</name>
<name>
<surname><![CDATA[Delgado-Saborit]]></surname>
<given-names><![CDATA[J.-M.]]></given-names>
</name>
<name>
<surname><![CDATA[Harrison]]></surname>
<given-names><![CDATA[R. M.]]></given-names>
</name>
<name>
<surname><![CDATA[Ruiz-Rudolph]]></surname>
<given-names><![CDATA[P.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Use of data imputation tools to reconstruct incomplete air quality datasets: A case-study in Temuco, Chile]]></article-title>
<source><![CDATA[Atmospheric Environment]]></source>
<year>2018</year>
</nlm-citation>
</ref>
<ref id="B7">
<label>(7)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Chen]]></surname>
<given-names><![CDATA[L.]]></given-names>
</name>
<name>
<surname><![CDATA[Wang]]></surname>
<given-names><![CDATA[G.]]></given-names>
</name>
<name>
<surname><![CDATA[Shen]]></surname>
<given-names><![CDATA[Z.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Comparison of the multiple imputation approaches for imputing rainfall data series and their applications to watershed models]]></article-title>
<source><![CDATA[Journal of Hydrology]]></source>
<year>2019</year>
</nlm-citation>
</ref>
<ref id="B8">
<label>(8)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Moritz]]></surname>
<given-names><![CDATA[S.]]></given-names>
</name>
<name>
<surname><![CDATA[Sardá]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
<name>
<surname><![CDATA[Bartz-Beielstein]]></surname>
<given-names><![CDATA[T.]]></given-names>
</name>
<name>
<surname><![CDATA[Zaefferer]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
<name>
<surname><![CDATA[Stork]]></surname>
<given-names><![CDATA[J.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Comparison of different Methods for Univariate Time Series Imputation in R]]></article-title>
<source><![CDATA[doi: 10.48550/arXiv.1510.03924]]></source>
<year>2015</year>
</nlm-citation>
</ref>
<ref id="B9">
<label>(9)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Cao]]></surname>
<given-names><![CDATA[W.]]></given-names>
</name>
<name>
<surname><![CDATA[Wang]]></surname>
<given-names><![CDATA[D.]]></given-names>
</name>
<name>
<surname><![CDATA[Zhou]]></surname>
<given-names><![CDATA[H.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[BRITS: Bidirectional Recurrent Imputation for Time Series]]></article-title>
<source><![CDATA[Neural Information Processing Systems 31]]></source>
<year>1805</year>
</nlm-citation>
</ref>
<ref id="B10">
<label>(10)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Oppong and S. Yao]]></surname>
<given-names><![CDATA[F.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Assessing Univariate and Multivariate Normality, A Guide For Non-Statisticians]]></article-title>
<source><![CDATA[Mathematical Theory and Modeling]]></source>
<year>2016</year>
</nlm-citation>
</ref>
<ref id="B11">
<label>(11)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Kim]]></surname>
<given-names><![CDATA[Y.]]></given-names>
</name>
<name>
<surname><![CDATA[Kim]]></surname>
<given-names><![CDATA[H.]]></given-names>
</name>
<name>
<surname><![CDATA[Lee]]></surname>
<given-names><![CDATA[G.]]></given-names>
</name>
<name>
<surname><![CDATA[Min]]></surname>
<given-names><![CDATA[K.-H.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[A Modified Hybrid Gamma and Generalized Pareto Distribution for Precipitation Data]]></article-title>
<source><![CDATA[Asia-Pacific Journal of Atmospheric Sciences]]></source>
<year>2019</year>
</nlm-citation>
</ref>
<ref id="B12">
<label>(12)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Mohammed]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[LSTM and Bidirectional LSTM for Regression - Towards Data Science]]></article-title>
<source><![CDATA[towardsdatascience.com/lstm-and-bidirectional-lstm-for-regression-4fddf910c655]]></source>
<year>2022</year>
</nlm-citation>
</ref>
<ref id="B13">
<label>(13)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Sucholutsky]]></surname>
<given-names><![CDATA[I.]]></given-names>
</name>
<name>
<surname><![CDATA[Narayan]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
<name>
<surname><![CDATA[Schonlau]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
<name>
<surname><![CDATA[Fischmeister]]></surname>
<given-names><![CDATA[S.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Deep Learning for System Trace Restoration]]></article-title>
<source><![CDATA[International Joint Conference on Neural Networks]]></source>
<year>1904</year>
</nlm-citation>
</ref>
<ref id="B14">
<label>(14)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Honaker]]></surname>
<given-names><![CDATA[J.]]></given-names>
</name>
<name>
<surname><![CDATA[King]]></surname>
<given-names><![CDATA[G.]]></given-names>
</name>
<name>
<surname><![CDATA[Blackwell]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Amelia II: A Program for Missing Data]]></article-title>
<source><![CDATA[Journal of Statistical Software]]></source>
<year>2011</year>
</nlm-citation>
</ref>
<ref id="B15">
<label>(15)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Miró]]></surname>
<given-names><![CDATA[J. J.]]></given-names>
</name>
<name>
<surname><![CDATA[Caselles]]></surname>
<given-names><![CDATA[V.]]></given-names>
</name>
<name>
<surname><![CDATA[Estrela]]></surname>
<given-names><![CDATA[M. J.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Multiple imputation of rainfall missing data in the Iberian Mediterranean context]]></article-title>
<source><![CDATA[Atmospheric Research]]></source>
<year>2017</year>
</nlm-citation>
</ref>
<ref id="B16">
<label>(16)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Desherevskii]]></surname>
<given-names><![CDATA[A. V.]]></given-names>
</name>
<name>
<surname><![CDATA[Zhuravlev]]></surname>
<given-names><![CDATA[I.]]></given-names>
</name>
<name>
<surname><![CDATA[Nikolsky]]></surname>
<given-names><![CDATA[N.]]></given-names>
</name>
<name>
<surname><![CDATA[Sidorin]]></surname>
<given-names><![CDATA[Y.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Problems in Analyzing Time Series with Gaps and Their Solution with the WinABD Software Package]]></article-title>
<source><![CDATA[Izvestiya, Atmospheric and Oceanic Physics]]></source>
<year>2018</year>
</nlm-citation>
</ref>
<ref id="B17">
<label>(17)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Andiojaya and H. Demirhan]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[A bagging algorithm for the imputation of missing values in time series]]></article-title>
<source><![CDATA[Expert Systems With Applications]]></source>
<year>2019</year>
</nlm-citation>
</ref>
<ref id="B18">
<label>(18)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Campozano]]></surname>
<given-names><![CDATA[L.]]></given-names>
</name>
<name>
<surname><![CDATA[Sanchez]]></surname>
<given-names><![CDATA[E.]]></given-names>
</name>
<name>
<surname><![CDATA[Avilés]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
<name>
<surname><![CDATA[Samaniego]]></surname>
<given-names><![CDATA[E.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Evaluation of infilling methods for time series of daily precipitation and temperature: The case of the Ecuadorian Andes]]></article-title>
<source><![CDATA[Maskana¸]]></source>
<year>2014</year>
</nlm-citation>
</ref>
<ref id="B19">
<label>(19)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Richman]]></surname>
<given-names><![CDATA[M. B.]]></given-names>
</name>
<name>
<surname><![CDATA[Trafalis]]></surname>
<given-names><![CDATA[T. B.]]></given-names>
</name>
<name>
<surname><![CDATA[Adrianto]]></surname>
<given-names><![CDATA[I.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Multiple imputation through machine learning algorithms]]></article-title>
<source><![CDATA[87th AMS Annual Meeting, 2007]]></source>
<year>2007</year>
</nlm-citation>
</ref>
<ref id="B20">
<label>(20)</label><nlm-citation citation-type="">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Zhai]]></surname>
<given-names><![CDATA[C.]]></given-names>
</name>
</person-group>
<source><![CDATA[&#8220;A Note on the Expectation-Maximization (EM) Algorithm&#8221;, 2004.]]></source>
<year>2004</year>
</nlm-citation>
</ref>
<ref id="B21">
<label>(21)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Honaker]]></surname>
<given-names><![CDATA[J.]]></given-names>
</name>
<name>
<surname><![CDATA[King]]></surname>
<given-names><![CDATA[G.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[What to do About Missing Values in Time Series Cross-Section Data]]></article-title>
<source><![CDATA[American Journal of Political Science]]></source>
<year>2010</year>
</nlm-citation>
</ref>
<ref id="B22">
<label>(22)</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Khampuengson and W. Wang]]></surname>
<given-names><![CDATA[T.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Novel Methods for Imputing Missing Values in Water Level Monitoring Data]]></article-title>
<source><![CDATA[Water Resources Management]]></source>
<year>2023</year>
</nlm-citation>
</ref>
</ref-list>
</back>
</article>
