BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//TYPO3/NONSGML News system (news)//EN
BEGIN:VEVENT
UID:news-8798@econ.au.dk
DTSTAMP:20260225T075510Z
DTSTART:20261111T111500Z
DTEND:20261111T123000Z
END:VEVENT
END:VCALENDAR




<div class="news news-single">
	<div class="article" itemscope="itemscope" itemtype="http://schema.org/Article">
		
	
			<script type="text/javascript">
				const showAllContentLangToken = "Show all content ";
			</script>

			
			

			<article class="typo3-delphinus delphinus-gutters">

				<!-- News PID: 17909 - used for finding folder/page which contains the news / event -->
				<!-- News UID: 8798 - the ID of the current news / event-->

				<div class="news-event">
					<div class="news-event__header">
						<!-- Categories -->
						
							<span class="text--stamp">
<!-- categories -->
<span class="news-list-category">
	
		
	
</span>

</span>
						

						<!-- Title -->
						<h1 itemprop="headline">Econometrics Seminar: Ines Wilms, Maastricht University</h1>
						
							<!-- Teaser -->
							<p class="text--intro" itemprop="description">Title: Cross-temporal forecast reconciliation using machine learning</p>
						
					</div>

					

					<div class="news-event__content">

						<!-- Events info box -->
						
								

								<div class="news-event__info theme--dark" id="event-info">
									<h2 class="screenreader-only">Info about event</h2>

									
											<!--- Same date -->
											<div class="news-event__info__item news-event__info__item--time">
												<h3 class="news-event__info__item__header text--label-header">Time</h3>
												<div class="news-event__info__item__content">
													<span class="u-avoid-wrap">
														Wednesday 11  November 2026,
													</span>
													<span class="u-avoid-wrap">
														&nbsp;at 12:15 -  13:30
													</span>
													<p class="news-event__info__item__ical-link"><a href="/about-us/news-and-events/single-event-item/artikel/econometrics-seminar-ines-wilms-maastricht-university?tx_news_pi1%5Bformat%5D=ical&amp;type=9819&amp;cHash=1907a42a64e4a2007f5bc46abe315491">Add to calendar</a></p>
												</div>
											</div>
										

									<!-- Location detailed -->
									
											<!-- Location Simple -->
											
												<div class="news-event__info__item">
													<h3 class="news-event__info__item__header text--label-header">Location</h3>
													<div class="news-event__info__item__content">
														<p>Universitetsbyen 51, Building 1816, room 613</p>
													</div>
												</div>
											
										

									<!-- Organizer detailed -->
									
											<!-- Organizer Simple -->
											
										

									<!-- Price -->
									

									<!-- Event link -->
									

									<!-- Registration -->
									
								</div>
							

						
							<!-- Media -->
							
								



							
						

						
							<div class="news-event__content__text">
								<span class="text--byline" id="byline">
									

									<!-- Author -->
									
										<span itemprop="author" itemscope="itemscope" itemtype="http://schema.org/Person">
											
													By
												

											
													<a href="mailto:sns@econ.au.dk">
														<span itemprop="name">Solveig Sørensen</span>
													</a>
												
										</span>
									
								</span>

								

									<!-- Body text -->
									<p>Speaker:&nbsp;Ines Wilms, Maastricht University (<a href="https://sites.google.com/view/iwilms" target="_self">website</a>)</p>
<p>Title: Cross-temporal forecast reconciliation using machine learning</p>
<p>Abstract: Many forecasting tasks involve multiple, interrelated time series that must satisfy linear aggregation constraints, where the components collectively sum to the total. Ensuring such coherence across all aggregation levels is the goal of forecast reconciliation, which is essential for consistent and aligned decision-making. In cross-temporal frameworks, the focus of this talk, these aggregation constraints extend across both cross-sectional and temporal dimensions. Existing literature primarily relies on linear reconciliation methods, which adjust base forecasts through linear transformations within a least-squares framework to satisfy aggregation constraints. In this work, we move beyond this paradigm and introduce a non-linear forecast reconciliation approach for cross-temporal frameworks. Our method directly and automatically produces cross-temporal coherent forecasts by leveraging popular machine learning techniques. We empirically validate our framework on large-scale streaming datasets from a leading European on-demand delivery platform and a bicycle-sharing system in New York City.</p>
<p>Host: Eric Hillebrand</p>
<hr>
<p>Organisers:&nbsp;<a href="https://econ.au.dk/contact/show/person/gsgudmundsson@econ.au.dk" target="_self">Stefán Guðmundsson</a>&nbsp;and <a href="https://econ.au.dk/contact/show/person/61e7210a-fc44-4a8b-9533-a056c20dd3e9" target="_self">Chen Huang</a></p><div><p>See all <a href="https://econ.au.dk/research/econometrics-and-business-analytics/seminars/econometrics-seminars" target="_self">Econometrics Seminars</a></p></div>
								
							</div>
						
					</div>

					
						<!-- Content elements -->
						
					
				</div>
			</article>

			
				
				
			

			<!-- related things -->
			
		

	</div>
</div>