Nowadays, a variety of multi-dimensional arrays (tensors) frequently occur in data analysis tasks.In this work it has been developed a new tool, TAJIVE, for integrative analysis which manages multidimensional data based on Tucker decomposition. Specifically, TAJIVE allows data reduction by expressing a large numberof observed variables by means of a smaller set of linear composites. It decomposes multi-block data into a sum of three components: a low rank approximation capturing joint structure between data types, low rank ap-proximations capturing structure individual to each data type, and residual noise. This tool is sufficiently generic to be used in different areas; in this work, TAJIVE has been applied to the energy markets with the intention of understanding the new behaviour of the ”conventional” market participants(CCGTs) after the arrival of the renewable energy systems.
Al giorno d’oggi, `e sempre pi`u frequente dover gestire nell’analisi dei datioggetti multidimensionali (tensori).In questo lavoro `e stato sviluppato un nuovo strumento per l’analisi integra-tiva basato sulla decomposizione di Tucker, TAJIVE, allo scopo di gestiredati multidimensionali. Nello specifico, TAJIVE decompone i dati multi-dimensionali nella somma di tre addendi: una componente che cattura lastruttura congiunta tra i datasets, una componente che incorpora la strut-tura individuale di ogni dataset e il rumore residuo. Questo strumento `e suf-ficientemente generico per essere utilizzato in diverse applicazioni; in questolavoro TAJIVE `e stato applicato ai mercati dell’energia con l’intento di com-prendere il comportamento dei partecipanti pi`u convenzionali (CCGTs) delmercato elettrico dopo l’arrivo dei sistemi di energia rinnovabile.
TAJIVE : a new tool for integrative analysis of multidimensional data. Application to the Italian Energy market
Marchionni, Virginie
2019/2020
Abstract
Nowadays, a variety of multi-dimensional arrays (tensors) frequently occur in data analysis tasks.In this work it has been developed a new tool, TAJIVE, for integrative analysis which manages multidimensional data based on Tucker decomposition. Specifically, TAJIVE allows data reduction by expressing a large numberof observed variables by means of a smaller set of linear composites. It decomposes multi-block data into a sum of three components: a low rank approximation capturing joint structure between data types, low rank ap-proximations capturing structure individual to each data type, and residual noise. This tool is sufficiently generic to be used in different areas; in this work, TAJIVE has been applied to the energy markets with the intention of understanding the new behaviour of the ”conventional” market participants(CCGTs) after the arrival of the renewable energy systems.File | Dimensione | Formato | |
---|---|---|---|
Tesi_marchionni.pdf
Open Access dal 28/11/2021
Dimensione
15.7 MB
Formato
Adobe PDF
|
15.7 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/169369