La tesi si compone di tre saggi su disoccupazione e mobilità del lavoro in Italia, presentando anche un focus sulla regione Lombardia, oltre che da una parte iniziale che inquadra tali tematiche. Il primo capitolo offre infatti una disamina degli sviluppi ed empirici connessi a disoccupazione e mobilità del lavoro. L’obiettivo di questa parte introduttiva è duplice. Da un lato si cerca di fornire un quadro pressochè esaustivo sulle evoluzioni teoriche ed empiriche connesse alle tematiche citate. D’altro lato si introducono le analisi oggetto dei successivi saggi come evoluzione degli sviluppi proposti dalla letteratura, enfatizzandone logiche sottostanti ed originalità. Il primo saggio analizza le determinanti della durata della disoccupazione ed i relativi “competing risks” per la regione Lombardia. La scelta di tale contesto non è casuale. La Lombardia, infatti, rappresenta una delle regioni economicamente più sviluppate ed i risultati ottenuti con tali metodologie di stima possono fornire spunti utili e rappresentativi sia delle regioni europee maggiormente sviluppate, sia di altre rilevanti regioni italiane (Emilia Romagna e Toscana). Il secondo saggio estende l’applicazione di modelli di durata e modelli a rischi competitivi all’intero territorio nazionale. In questo modo è possibile enfatizzare la rilevanza di tali tematiche per il contesto italiano, ed ottenere un quadro esaustivo circa l’evoluzione del fenomeno della durata della disoccupazione. Le tecniche utilizzate per tali analisi, ovviamente, differiscono ripetto a quelle impegate per la regione Lombardia, ed anche questo aspetto consente interessanti considerazioni. Il terzo saggio sposta l’attenzione alla rilevante tematica della mobilità del mercato del lavoro. Tale aspetto è ovviamente connesso al fenomeno della disoccupazione, e consente di approfondirne nonché di delinearne le possibili cause. In tale capitolo vengono proposte due metodologie di analisi. In primo luogo, ed a livello macro, sono fornite le stime aggregate dei flussi fra i principali stati o condizioni (occupazione, disoccupazione, inattività) del mercato del lavoro. Questo primo step consente appunto una prima quantificazione del fenomeno della mobilità. La seconda parte del capitolo si focalizza invece su una stima - a livello micro - delle determinanti delle transizioni fra gli stati del mercato del lavoro. Tale aspetto consente appunto di investigare ed esaminare le cause sottese alla mobilità riscontrata a livello macro. / Structured in three essays, this thesis focus on unemployment and labour mobility in Italy and Lombardy (the biggest Italian’s region). The first essay offers a picture of the main theoretical and the empirical issues related to these complex phenomena. The purpose of this section is twofold. On one hand we aim to offer an exhaustive picture of the theoretical and empirical developments of such phenomena. On the other hand, we introduce the empirical investigations of the subsequent essays as evolutions of the ones proposed by literature. We also emphases the original contribution and the logic behind. The second essay investigates the determinants of the unemployment duration and of the related competing risks (CRM hereafter) for Lombardy. The choice to concentrate the initial part of this dissertation on Lombardy is primarily driven by two factors. First, there is interest in applying relevant techniques to a regional context characterized by a certain degree of homogeneity of economic indicators. Further, Lombardy is one of the most important Italian regions (confirmed by many economics indicators), and is quite homogeneous in terms of labour market indicators (only little differences between provinces, with the north-east with the fewest unemployment problems), This allows verifying the effectiveness of these investigations of the determinants of unemployment duration and the related CRM without dealing with the typical dualism between north and south which is a structural feature of the Italian labour market. This is a way to investigate in depth the characteristics of the relevant phenomenon of unemployment for a significant partition of Italy, which is representative of both richest regions in Europe and Italian regions as well (such as Tuscany or Emilia Romagna). The third essay enlarges the attention to Italy by employing techniques of unemployment duration and competing risks to analyse the overall Italian unemployment and its main exit routes. Those are tools to get an exhaustive picture and relevant insights on the evolution of the Italian unemployment duration. The techniques employed for the overall country obviously differ from the ones used for the region of Lombardy, and these differences also offer the scope for interesting considerations. The fourth essay deals with the relevant issue of labour market mobility. This is a theme quite linked to unemployment, since it allows understanding and exploring its causes. We focus on two different kind of analysis. At macro level, we estimate the gross flows between the relevant labour market states of employment, unemployment, and inactivity (three-state representation of the labour market) to quantify the overall labour market mobility. The second part of this section, instead, offers micro econometrics estimates of the determinants of such labour market transitions, to investigate the causes of such mobility.

Value-items from the Schwartz scale of Values have been added to travel data to investigate if the value-items can be used to model mode choice. Two kinds of mode choice models, both discrete choice models, multinomial models (MNL) and the Machine Learning Models Random Forests (RF) were constructed, using Travel Diary data (RVU) and additional data from European Social Survey (ESS). The additional data was connected to the base data by grouping the individuals using three key variables: gender, age, and household income. Models were then created with and without any data from the value-items to screen for variables that had an impact on the model. The RF model predicted the correct modes for all but the smaller groups, car passengers and biking. While the MNL model had less success accurately assessing which mode someone had chosen. The MNL with additional grouped value-items improved, while the models created using Random Forest had no difference in accuracy based on the addition. Even though there were some significant value-items in the MNL-models, the expected consequences from them small, as the base model specification might be insufficient in incorporating more relevant variables. Based on the Random Forest having no use from the value-items along with them being of similar importance no value-items stood out for further testing. The main findings were thus that no value-items of particular interest could be found with the RF model while the results for the MNL-model were inconclusive. Suggested improvements for further similar studies would be to perform grouping using data for a longer time frame and or to use a value-model as input for the mode choice modelling. It is deemed appropriate to study what values people associate with specific modes directly, and to investigate if other models such as car ownership models or models of choices between different versions of the same mode could be more suitable for additional value-data. / Värdeobjekt från Schwartz värderingsskala har kombinerats med resedata för att undersöka om värdeobjekten kan användas vid modellering av färdmedelsval. Två typer av färdmedelsmodeller, multinomiala modeller (MNL) och Random Forests konstruerades. Den data som användes var Resvanedata (RVU), med kompletterande värderingsdata från European Social Survey (ESS). ESS-datan kopplades till basdatan genom att gruppera individerna med hjälp av tre nyckelvariabler: kön, ålder och hushållsinkomst. Sedan skapades modeller med och utan den kompletterande datan för att se om modellerna påverkades. RF-modellens resultat överensstämde väl med de faktiska valen förutom för de mindre grupperna: bilpassagerare och cyklister. MNL-modellen hade mindre framgång med att bedöma vilket färdmedel en individ hade valt. MNL-modellen med ytterligare grupperade värdeobjekt förbättrades i jämförelse med grundmodellen, medan modellerna skapade med Random Forest inte skilde sig märkbart från varandra. Även om värdeobjekten i MNL-modellerna var signifikanta är de förväntade konsekvenserna av dem små, eftersom specifikationen för basmodellen tros saknar andra mer relevanta variabler. RF-modellen gynnades inte av värdeobjekten och inga värdeobjekt var betydande för modellen. De huvudsakliga fynden var att inga värdeobjekt av särskilt intresse kunde hittas med RF-modellen medan resultaten för MNL-modellen var ofullständiga. Föreslagna förbättringar för ytterligare liknande studier skulle vara att utföra gruppering med hjälp av data för ett längre tidsspann eller att introducera en värdemodell som indata för modelleringen av färdmedelsval. Det bedöms lämpligt att studera vilka värderingar människor förknippar med specifika färdmedel direkt samt att undersöka om andra modeller såsom av bilägande eller i val mellan olika versioner av samma färdmedel skulle var mer passande för att modelleras med hjälp av data med värderingar.