Σχολή Οικονομικών Επιστημών
Μόνιμο URI για αυτήν την κοινότηταhttps://pyxida.aueb.gr/handle/123456789/58
Η Σχολή Οικονομικών Επιστημών περιλαμβάνει τα Τμήματα: - Τμήμα Οικονομικής Επιστήμης - Τμήμα Διεθνών και Ευρωπαϊκών Οικονομικών Σπουδών.
Περιήγηση
Πλοήγηση Σχολή Οικονομικών Επιστημών ανά Τίτλο
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Α Β Γ Δ Ε Ζ Η Θ Ι Κ Λ Μ Ν Ξ Ο Π Ρ Σ Τ Υ Φ Χ Ψ Ω
Τώρα δείχνει 1 - 20 από 2811
- Αποτελέσματα ανά σελίδα
- Επιλογές ταξινόμησης
Τεκμήριο The 0.5 global Sulphur cap and its implications on the tanker industryKornilaki, Anastasia; Nikolaou, Ioanna; Athens University of Economics and Business, Department of Accounting and Finance; Athens University of Economics and Business, Department of International European Economic Studies; Panayides, PhotisThe dissertation investigates the available methods with which shipowners can comply with the Sulfur regulations of bunker fuels that will be enforced in 2020. To begin with, general characteristics of the tanker shipping industry are presented, such as oil and oil products, ships carrying oil at sea, as well as sources of marine pollution. In continuation, a deep literature review is being carried out on the regulations that have been or will be adopted regarding Sulfur limit and emissions from ships’ engines. We also identify the sea areas to which stricter regulations apply, namely Emissions Control Areas (ECAs) and Particularly Sensitive Sea Areas (PSSAs). The factors that shipowners need to take into consideration and can affect their decision concerning the choice of the optimal method of compliance, are also examined. Furthermore, the available methods with which hipowner can meet the new sulfur requirements, together with the advantages and disadvantages of each method are being analyzed. Finally, a comparison between the various alternative methods is being performed using market reports, previous researches on that subject, a SWOT analysis, a cost comparison of the available methods for three tankers with different characteristics and an ECA calculator in order to calculate the payback period of the proposed investment.Τεκμήριο 1. Handling data from Thompson’s DataStream 2. The Cross-Section of European Returns: an application of asset pricing models, anomalies and predictability in a Pan-European frameworkLandis-Conrad, Felix-Michel; Athens University of Economics and Business, Department of International European Economic Studies; Skouras, SpyrosThis paper refers on Handling data from Thompson’s DataStream (TDS). Although DataStream is a rich data source for approximately 50.000 equities from more than 65 markets around the world with more than 25 years of coverage, its data should be handle with care. An extensive research on that subject is already published by Ince and Porter (2006-“Individual Equity Return Data from Thompson DataStream: Handle With Care”). So our paper comes as a supplement to it, providing, though more information especially in the common stock identification processes for European markets and proposing extra screens for unreported errors. At the same time it, jointly describes the methodology we applied to obtain the European sample we used in our tests throughout the paper 2 .Τεκμήριο 1997 Asian crisis - the vulnerabilities of the "Asian Tigers" and the fiscal response towards recovery: a country by country analysis(2019-02-18) Kottas, Christos; Athens University of Economics and Business, Department of International and European Economic Studies; Tsakloglou, Panagiotis; Moutos, ThomasThe Asian economic and financial crisis that occurred in 1997 is considered as one of the most serious crises in the economic history. The contagion effect, following the depreciation of the Thai baht due to the massive speculative attack on July 1997, transmitted the crisis in the other countries of the area, mainly South Korea, Malaysia, Philippines, Singapore and Indonesia. This paper examines the anatomy of the East Asian financial crisis, analyzes the fiscal response and the measures that were implemented and assesses the macroeconomic adjustment of the involved countries by presenting the results of the measures implemented and its output the years afterwards. The analysis conducted is firstly country – oriented, that is, each country’s involvement in the crisis is examined separately focusing on the macroeconomic framework before the crisis. Moving towards the fiscal adjustment section of the paper, the analysis remains on a country-by-country level in order to analyze the measures implemented aiming at the assessment of the supporting programs degree of success and focusing on the International Monetary Fund’s and countries authorities intervention there, as well as the fiscal adjustments modalities that became necessary in this context. Finally, the paper overviews the situation and aftermath of these countries the years afterwards and the position of these countries in the modern world almost 20 years after the end of the Asian tigers crisis.Τεκμήριο A combination of machine learning with classical econometric approaches for temporal disaggregationΒαρελάς, Κωνσταντίνος; Varelas, Konstantinos; Athens University of Economics and Business, Department of Economics; Antoniou, Fabio; Pagratis, Spyros; Alexopoulos, AngelosΑυτή η μελέτη συνεισφέρει στη βιβλιογραφία του χρονικού επιμερισμού σειρών στο πλαίσιο της οικονομίας και ειδικότερα στο πλαίσιο της άυλης οικονομίας. Μέχρι τώρα η έρευνα επικεντρωνόταν κυρίως σε στατικές και δυναμικές παλινδρομήσεις και σε διαρθρωτικά υποδείγματα χρονολογικών σειρών, όμως αυτές οι προσεγγίσεις συνεπάγονται κάποιες υποθέσεις. Σε αυτή τη διπλωματική προτείνεται μία υβριδική μέθοδος που συνδυάζει XGBoost και GLS μοντέλα. Τα οφέλη αυτής της προσέγγισης είναι η μη ύπαρξη υποθέσεων, η δυνατότητα χρήσης ενός εξαντλητικού αριθμού ανεξάρτητων μεταβλητών και η συνέπεια στα αποτελέσματα. Συνολικά 33 μοντέλα αξιολογούνται βάσει ποσοτικών και ποιοτικών κριτηρίων σε δεδομένα που αφορούν 16 ευρωπαϊκές χώρες και διάφορες χρονικές περιόδους. Επίσης, προσπάθεια γίνεται να αναπαραχθούν τα αποτελέσματα των Αλεξόπουλου και Βαρθαλίτη που ήταν οι πρώτοι που χρησιμοποίησαν XGBoost για να επιμερίσουν χρονολογικές σειρές. Τέλος, οι σειρές GVA adjusted, Non-National Accounts Intangible Investment, and Total Intangible Investment επιμερίζονται χρονικά από ετήσια σε τριμηνιαία συχνότητα για πρώτη φορά. Ο επιμερισμός καλύπτει 19 επιμέρους τομείς της οικονομίας συμπεριλαμβανομένων των συνόλων των τομέων 16 ευρωπαϊκών χωρών για κάθε μία από τις 3 προαναφερθείσες μεταβλητές. Τα δεδομένα αντλούνται από τη βάση EUKLEMS & INTANProd και την Eurostat.Τεκμήριο A comprehensive survey of economic and financial time series forecasting: from econometric models to PCA-based statistical arbitrage(2026-01-19) Vavoule, Pinelopi; Βαβουλέ, Πηνελόπη; Pagratis, Spyridon; Kospentaris, Ioannis; Tzavalis, EliasΗ παρούσα διατριβή διερευνά την εξέλιξη και την εμπειρική εφαρμογή των μεθοδολογιών πρόβλεψης σε οικονομικές και χρηματοοικονομικές χρονοσειρές. Εξετάζει τη μετάβαση από τα κλασικά οικονομετρικά υποδείγματα – όπως τα ARIMA, GARCH και VAR – προς τα σύγχρονα πλαίσια μηχανικής μάθησης, συμπεριλαμβανομένων των Gradient Boosting και Transformer. Η ανάλυση αναδεικνύει τους συμβιβασμούς μεταξύ ερμηνευσιμότητας και προγνωστικής ακρίβειας, καθώς και τον αυξανόμενο ρόλο των δεδομενοκεντρικών προσεγγίσεων σε περιβάλλοντα υψηλής διαστατικότητας. Κεντρική συμβολή της μελέτης αποτελεί η ενσωμάτωση της Ανάλυσης Κύριων Συνιστωσών (PCA) ως ενοποιητικού στατιστικού πλαισίου που γεφυρώνει την οικονομετρία με τις υπολογιστικές μεθόδους. Η PCA εφαρμόζεται σε εκτεταμένα πάνελ αποδόσεων μετοχών για την εξαγωγή λανθανόντων παραγόντων που αποτυπώνουν τη συστηματική δυναμική των αγορών, οι οποίοι στη συνέχεια ενσωματώνονται σε κανονικοποιημένα παλινδρομικά μοντέλα για την πρόβλεψη μεταβολών στην καμπύλη αποδόσεων των αμερικανικών κρατικών ομολόγων. Η εμπειρική ανάλυση βασίζεται σε ημερήσια δεδομένα τιμών μετοχών των 503 εταιρειών του δείκτη S&P 500 από το Yahoo Finance και στη διαφορά αποδόσεων 10ετούς – 2ετούς διάρκειας των αμερικανικών ομολόγων από τη βάση FRED για την περίοδο Σεπτέμβριος 2020 – Σεπτέμβριος 2025. Τα αποτελέσματα δείχνουν ότι τα υποδείγματα PCA, ιδιαίτερα εκείνα με κανονικοποίηση LASSO, υπερέχουν των κλασικών οικονομετρικών μοντέλων ως προς την ακρίβεια και τη σταθερότητα πρόβλεψης. Συνολικά, τα ευρήματα υποστηρίζουν τις υβριδικές προσεγγίσεις που συνδυάζουν την οικονομετρική δομή με τις δεδομενο-κεντρικές μεθόδους ως αποτελεσματικό πλαίσιο για την κατανόηση και πρόβλεψη των πολύπλοκων δυναμικών των χρηματοοικονομικών αγορών.Τεκμήριο A simulation-based comparison of pseudo-marginal MCMC methods for the stochastic volatility model(2026-01-28) Gevorgyan, Hayk; Tzavalis, Elias; Dendramis, Yiannis; Alexopoulos, AngelosStochastic volatility (SV) models offer a flexible framework to capture the time-varying nature of the volatility process, but their complex structure, which stems from the intractability of the likelihood function, makes parameter inference challenging. The dissertation provides a detailed, simulation-based comparison of several advanced Bayesian estimation methods designed to overcome this challenge, focusing on the family of pseudo-marginal Markov chain Monte Carlo (MCMC) algorithms. The core of this research is a simulation study that evaluates the performance of four main algorithms: the pseudo-marginal algorithm with importance sampler (PM-IS), its correlated version (CPM-IS), the Particle Marginal Metropolis-Hastings (PMMH) algorithm, and its correlated version (CPMMH). Two different proposal covariance structures are used for each algorithm variation - a simple diagonal covariance and an adaptive covariance. The algorithms are tested on simulated datasets of two sample sizes (𝑇 = 200 and 𝑇 = 700). The evaluation is conducted in a two-phase process. First, the reliable posterior convergence of all algorithm variants is evaluated by conducting convergence diagnostics. Second, the methods that pass these first-step checks are compared on their computational efficiency and statistical accuracy. The convergence diagnostics reveal that the choice of both the likelihood estimator and the proposal mechanism is essential for reliable inference. Only the PMMH variants that utilized an adaptive proposal covariance mechanism consistently passed all diagnostic checks, demonstrating the necessity of a proposal that can learn the structure of the correlations between parameters. The findings lead to a clear and practical recommendation: for robust and efficient estimation of SV models of relatively small sample sizes (𝑇 ≤ 700), the non-correlated Particle Marginal Metropolis-Hastings (PMMH) algorithm with an adaptive proposal mechanism is the most effective choice.Τεκμήριο A structure analysis and the econometrics of cryptocurrencies(2022) Στάικος, Θεόδωρος; Οικονομικό Πανεπιστήμιο Αθηνών, Τμήμα Οικονομικής Επιστήμης; Παγκράτης, Σπυρίδων; Δενδραμής, Ιωάννης; Τζαβαλής, ΗλίαςΜε τη πάροδο του χρόνου είναι εμφανές πως υπάρχει μια συνεχής αύξηση των χρηματιστηριακών προϊόντων, όπως τα Futures. Options, CDS και ως τελευταία προσθήκη η είσοδος των κρυπτονομισμάτων. Ιστορικά, το χρηματοπιστωτικό σύστημα πέρασε από κύματα αναταράξεων με έντονη νομισματική αστάθεια λόγω υπερπληθωριστικών τάσεων και μεγάλων διακυμάνσεων στις συναλλαγματικές ισοτιμίες. Στο κοντινό παρελθόν το 2008 και στα χρόνια της παγκόσμιας οικονομικής κρίσης, χώρες ανά τον κόσμο, μέσα σε αυτές και η Ελλάδα, αντιμετώπιζαν τεράστια ελλείματα στον κρατικό προϋπολογισμό και το ισοζύγιο τρεχουσών συναλλαγών δημιουργώντας ένα αίσθημα νευρικότητας και έρπουσας ανησυχίας για το παρόν και κυρίως το μέλλον του διεθνούς νομισματικού συστήματος. Η είσοδος του Bitcoin το 2008 από ένα άτομο ή μία ομάδα προγραμματιστών με το ψευδώνυμο “Satoshi Nakamoto” παρουσίασε μία νέα πτυχή πραγματοποίησης συναλλαγών αποκομμένη από το διατραπεζικό σύστημα, βασισμένο στη κρυπτογραφία και σε ορισμένα πρωτόκολλα τα οποία θα αναλυθούν εκτενώς στην συνέχεια, προσφέροντας εύφορο έδαφος στην δημιουργία, με γεωμετρική πρόοδο θα μπορούσε να ειπωθεί, νέων κρυπτονομισμάτων έως σήμερα. Οι αποκεντρωμένες συναλλαγές παρά τη φαινομενικά απόλυτη διαφάνειά τους και την ευκολία τους στην εκτέλεση, διεγείρουν σημαντικά ερωτήματα στην παγκόσμια οικονομική κοινότητα. Μπορεί να υπάρξει κορεσμός στην αγορά κρυπτονομισμάτων; Μπορούν να περιοριστούν οι συναλλαγές παράνομης δραστηριότητας μέσω ενός κανονιστικού πλαισίου; Δύναται τα κρυπτονομίσματα να αντικαταστήσουν τα παραδοσιακά νομίσματα; Είναι ενδιαφέρον ως επενδυτικό προϊόν τα κρυπτονομίσματα ή είναι μια τεράστια «φούσκα» που θα σκάσει στο εγγύς μέλλον; Ποιοι είναι οι κίνδυνοι που ελλοχεύουν οι επενδύσεις σε κρυπτονομίσματα και ποια η σχέση τους με τα παραδοσιακά assets; Θα αναλυθεί επίσης η λειτουργία των κρυπτονομισμάτων, κατανοώντας τη φύση τους και θα μελετηθεί η διακύμανση των χρονοσειρών τους, καθώς και το αν, και σε ποιο βαθμό, επηρεάζεται η αγορά κρυπτονομισμάτων από την covid-19 περίοδο που διανύουμε.Τεκμήριο Active portfolio management with applications(Athens University of Economics and Business, 2007-12) Papadakis, Dimitrios; Kordas, GregoryThesis - Athens University of Economics and Business. Postgraduate, Department of EconomicsΤεκμήριο Adaptation of OECD anti corruption guidelines: a study of selected MNCs(2024-03-27) Kiritsi, Anastasia; Κυρίτση, Αναστασία; Athens University of Economics and Business, Department of International and European Economic Studies; Katsimi, Margarita; Topaloglou, Nikolaos; Moutos, ThomasΑυτή η μελέτη εξετάζει τις προσπάθειες καταπολέμησης της διαφθοράς των πολυεθνικών εταιρειών για να ανακαλύψουν πόσο δύσκολες είναι, πόσο περίπλοκες είναι και πώς να τις κάνουν πιο ανοιχτές. Μια διεξοδική μελέτη της βιβλιογραφίας και των δεδομένων βρίσκει σημαντικά θέματα και προσφέρει τρόπους για τις πολυεθνικές εταιρείες να βελτιώσουν τις προσπάθειές τους για την καταπολέμηση της διαφθοράς. Η μελέτη αναφέρει ότι οι πολυεθνικές εταιρείες πρέπει να γνωρίζουν και να ακολουθούν τους κανόνες κατά της διαφθοράς. Η κατανόηση των συμμετεχόντων επηρεάζεται από το πού εργάζονται και τι έχουν εκπαιδευτεί. Η μελέτη δείχνει επίσης ότι είναι δύσκολο να τηρηθούν οι κανόνες λόγω των περιορισμένων πόρων, της αντίστασης των εργαζομένων και της δυσκολίας των κανόνων. Παρά τα προβλήματα αυτά, η μελέτη τονίζει τη σημασία ισχυρών μέτρων κατά της διαφθοράς για καλή ηθική και κυβέρνηση. Όπως είδαμε, η ηθική και το καθήκον επηρεάζονται από τη νοοτροπία μιας εταιρείας και την αφοσίωση των ηγετών της. Η μελέτη διερευνά εάν οι χώρες μπορούν να συνεργαστούν για την καταπολέμηση του εγκλήματος και εξετάζει πώς επηρεάζει τις επιχειρήσεις και την οικονομική ανάπτυξη σε όλο τον κόσμο. Αυτή η μελέτη δίνει πολλές ιδέες για το πώς οι πολυεθνικές εταιρείες μπορούν να καταπολεμήσουν το μόσχευμα. Ορισμένες από αυτές υποστηρίζουν την ηθική, διασφαλίζουν ότι οι κανόνες ακολουθούνται καλύτερα και χρησιμοποιούν την τεχνολογία. Προκειμένου να καταπολεμηθεί η διαφθορά, η δωροδοκία και η νομιμοποίηση εσόδων από παράνομες δραστηριότητες, η μελέτη επικεντρώνεται σε μέτρα πολιτικής όπως η παγκόσμια κυβέρνηση και η συνεργασία με άλλες χώρες.Τεκμήριο Adjusting portfolio value at risk using options(Οικονομικό Πανεπιστήμιο Αθηνών, 2010-11) Πλασσαράς, Παναγιώτης; Οικονομικό Πανεπιστήμιο Αθηνών, Τμήμα Οικονομικής Επιστήμης; Κόρδας, ΓρηγόρηςΔιπλωματική εργασία - Οικονομικό Πανεπιστήμιο Αθηνών. ΜΠΣ στα Εφαρμοσμένα Οικονομικά και ΧρηματοοικονομικάΤεκμήριο Advanced modeling in sports analytics(2026-02-04) Motsios, Evangelos; Μότσιος, Ευάγγελος; Pagratis, Spyridon; Zacharias, Eleftherios; Alexopoulos, AngelosThis master’s thesis investigates the development, evaluation, and interpretation of advanced Expected Goals (xG) models in football, using event-level data and modern machine learning techniques. The main objective is to examine whether different methodological approaches, ranging from simple statistical models to more complex algorithmic methods, can provide reliable, well-calibrated, and interpretable probabilistic estimates of scoring probability. The analysis is based on StatsBomb Open Data and includes all shots from the five major European leagues (England, Spain, Italy, Germany, and France) during the 2015/2016 season. The final dataset consists of more than 1,800 matches and tens of thousands of shots, enabling shot-level modeling of goal probability. The methodology follows a fully reproducible analytical pipeline, including data collection and cleaning, feature engineering, train–test splitting, model training, evaluation, probability calibration, and model interpretation. Logistic regression is used as a baseline model, while tree-based methods (decision trees and random forests) and boosting models (XGBoost) are also examined. Model performance is evaluated using metrics suitable for probabilistic prediction under strong class imbalance, such as the Area Under the ROC Curve (AUC) and the Brier score, rather than traditional classification metrics. Particular emphasis is placed on probability calibration, employing methods such as Platt scaling and isotonic regression. The results indicate that boosting models achieve superior discriminative performance compared to simpler approaches; however, they require careful calibration to ensure reliable probability estimates. Furthermore, the application of techniques such as the SHAP values enables both global and local interpretation of model predictions, highlighting the dominant role of spatial and contextual factors such as shot distance, angle, and shooting conditions. Overall, the thesis demonstrates that xG modeling is not merely a prediction task but a comprehensive methodological problem that requires the integration of appropriate data, evaluation strategies, calibration procedures, and interpretability tools. The findings underline the value of xG models for both academic research and practical football analysis, while also outlining future research directions including causal inference, phase-specific xG modeling, and the integration of richer data sources.Τεκμήριο Advanced regression analysis for trading and investment with an application to the EUR/US exchange rate(Οικονομικό Πανεπιστήμιο Αθηνών, 2007-12) Maratos, Andreas; Κόρδας, ΓρηγόρηςΔιπλωματική εργασία - Οικονομικό Πανεπιστήμιο Αθηνών. ΜΠΣ, Τμήμα Οικονομικής ΕπιστήμηςΤεκμήριο Aggregate and distributional implications of fiscal consolidation in general equilibrium(2019) Roumpanis, Eleftherios Theodoros; Ρουμπάνης, Ελευθέριος Θεόδωρος; Athens University of Economics and Business, Department of Economics; Σακελλάρης, Πλούταρχος; Βασιλάτος, Ευάγγελος; Τζαβαλής, Ηλίας; Οικονομίδης, Γεώργιος; Κωνσταντίνου, Παναγιώτης; Διοικητόπουλος, Ευάγγελος; Φιλιππόπουλος, ΑπόστολοςThis thesis is about the aggregate and distributional implications of fiscal consolidation in New Keynesian D(S)GE models. The thesis studies how these implications depend on the specific fiscal policy instrument used for debt consolidation. Chapter 2 presents a closed-economy New Keynesian D(S)GE model. Chapter 3 extends the model of Chapter 2 to set up a New Keynesian model of a small open economy within a monetary union facing sovereign interest rate premia. Finally, Chapter 4 builds a New Keynesian D(S)GE model consisting of two heterogeneous countries participating in a monetary union.Τεκμήριο Aggregates vs disaggregates using machine learning(2022-03-11) Μνηματίδου, Αναστασία; Οικονομικό Πανεπιστήμιο Αθηνών, Τμήμα Οικονομικής Επιστήμης; Τζαβαλής, Ηλίας; Δενδραμής, Ιωάννης; Παπαηλίας, ΦώτηςΜία μέθοδος για τον υπολογισμό άμεσων προβλέψεων για επίσημους δείκτες τιμών είναι η χρήση τιμών που είναι διαθέσιμες στο διαδίκτυο. Αυτή η ιδέα ξεκίνησε από τον Cavallo (2013) που χρησιμοποίησε διαδικτυακούς δείκτες τιμών για να επαληθεύσει τα επίσημα στοιχεία που ανακοινώθηκαν για τον πληθωρισμό. Ωστόσο, ο Cavallo χρησιμοποιεί το συνολικό άθροισμα των πληροφοριών σχετικά με τις τιμές που είναι διαθέσιμες στο διαδίκτυο για να υπολογίσει τον Δείκτη Τιμών Καταναλωτή. Αυτή η εργασία διερευνά εάν οι άμεσες προβλέψεις των επίσημων δεικτών τιμών είναι πιο ακριβείς όταν χρησιμοποιούνται οι επιμέρους ομάδες αγαθών και υπηρεσιών (main aggregates) και οι επιμέρους υποκατηγορίες (disaggregates). Συμπεραίνει ότι όλες οι μέθοδοι πρέπει να λαμβάνονται υπόψη διότι σε πολλές περιπτώσεις τα main aggregates και τα disaggregates υπερέχουν στην πραγματικότητα από το total aggregate.Τεκμήριο The agricultural productivity in Sub-Saharan Africa: an application of non-linear panel data modelsKonstantinou, Adamantia; Athens University of Economics and Business, Department of Economics; Kyriazidou, EkateriniDespite the fact that sub-Saharan countries are characterised by high potentials in the agricultural sector, it seems that they do not exploit them. Even if many researchers recommend an increase in the use of chemical fertilizers so that higher agricultural productivity rates could be achieved, Africa is still behind in fertilizer use. In this dissertation, we try to explore the factors which determine the adoption and intensity of chemical fertilizer use. The analysis is based on a panel dataset from the Amhara region of Ethiopia. Both the farm characteristics and the socio-economic ones of the rural households are essential for taking the decision about the adoption and fertilizer use. The positive effects from the participation in MFIs and cooperatives and the negative effects from the credit constraints on the adoption, indicate the important role which the structure of rural credit markets and the access on credits play on the use of chemical fertilizers.Τεκμήριο AI-driven profits by AI pricing methods. Collusion evidence in the digital business world(2023-02-28) Ζάχαρη, Σπυριδούλα; Zachari, Spyridoula; Athens University of Economics and Business, Department of Economics; Antoniou, Fabio; Dioikitopoulos, Evangelos; Zacharias, EleftheriosΗ εμφάνιση της τεχνητής νοημοσύνης (ΤΝ) έχει αλλάξει τη βιομηχανία και έχει δημιουργήσει μια νέα δυναμική σε ό,τι αφορά τις ευκαιρίες, την αποτελεσματικότητα και την ακρίβεια. Οι ψηφιακές αγορές έχουν ήδη εκμεταλλευτεί τις δυνατότητες που προσφέρει σε όλους τους τομείς και την έχουν ενσωματώσει σε διαδικασίες δυναμικής τιμολόγησης. Η ΤΝ, ως τμήμα της Μηχανικής Μάθησης, αναπτύσσεται σύμφωνα με τις πιο πρόσφατες τεχνολογικές καινοτομίες. Έτσι, οι εταιρείες που χρησιμοποιούν αλγόριθμους τιμολόγησης με χρήση ενισχυτικής μάθησης στην ψηφιακή αγορά, γίνονται πιο αποτελεσματικές όσον αφορά τις γρήγορες προσαρμογές των τιμών. Πρόσφατες επιστημονικές μελέτες, τόσο στα οικονομικά όσο και στα νομικά, τόνισαν τις πιθανές στρεβλώσεις του ανταγωνισμού και πιο συγκεκριμένα εστιάζουν στο πρόβλημα της αλγοριθμικής σιωπηρής συμπαιγνίας λόγω της αυτόνομης συμπεριφοράς των αλγορίθμων τιμολόγησης. Σκοπός αυτής της διατριβής είναι να εξετάσει την εφαρμογή της τεχνητής νοημοσύνης στα λογισμικά τιμολόγησης και να αναδείξει την πιο πρόσφατη βιβλιογραφία που επικεντρώνεται στις στρεβλώσεις του ανταγωνισμού από τη χρήση αλγορίθμων τιμολόγησης με ΤΝ. Το πρώτο κεφάλαιο διευκρινίζει το υπό συζήτηση ζήτημα σκιαγραφώντας τα τεχνικά στοιχεία που πλαισιώνουν την αλγοριθμική τιμολόγηση. Το δεύτερο κεφάλαιο εστιάζει στον τομέα της τιμολόγησης ως διακριτό τομέα της βιομηχανικής οργάνωσης. Το τρίτο κεφάλαιο παρουσιάζει τους θεωρητικούς κινδύνους που συνοδεύουν τις μεθόδους ML και μια ολοκληρωμένη ανάλυση των υποθέσεων για τις επιπτώσεις στον ανταγωνισμό όταν οι ψηφιακές αγορές χρησιμοποιούν αλγοριθμική τιμολόγηση. Στο κεφάλαιο 4 παρουσιάζονται οι μηχανισμοί πίσω από την αλγοριθμική σιωπηρή συμπαιγνία και οι αγορές που είναι πιο επιρρεπείς σε αυτές τις πρακτικές. Στο κεφάλαιο 5 η βιβλιογραφική ανασκόπηση επιχειρεί να παρουσιάσει μια ενδελεχή συλλογή των πιο σημαντικών επιστημονικών papers που αντιμετωπίζουν το πρόβλημα του αλγοριθμικού συντονισμού χρησιμοποιώντας εξελιγμένους αλγόριθμους. Το κεφάλαιο 6 παρουσιάζει το πλαίσιο της διεθνούς πολιτικής για την τεχνητή νοημοσύνη και το ζήτημα της ηθικής που ανακύπτει. Τέλος, η τελευταία ενότητα είναι αφιερωμένη στην παρουσίαση των συμπερασμάτων της προαναφερθείσας εκτενούς ανάλυσης.Τεκμήριο Aid for economic development: the efficiency of aid to developing countries, the case of Africa(2021) Kampouri, Anna-Kanelia; Καμπούρη, Άννα-Κανέλια; Athens University of Economics and Business, Department of International European Economic Studies; Milliou, Chrysovalantou; Chatzipanagiotou, PanagiotisForeign aid is a great topic of discussion between developed and developing world the last 50 years. This thesis is separated in two main parts. The first one concerns the theory of foreign aid, the effects and motives of the donors, while the second part analyzes the case of Africa. It is worth noting that while Africa has received huge amounts of foreign aid still remains in a very low level of livelihood. The impressive characteristic is that countries, which receive the aid, tend to be rich in natural resources, however those countries are underdeveloped and live in poverty with corrupted health and education systems. However, it seems to be an economic growth in the very last years in some countries such as Ethiopia.Τεκμήριο Aid for economic development: the efficiency of EU aid programs to developing countriesKoukouras, Konstantinos A.; Athens University of Economics and Business, Department of International European Economic Studies; Hatzipanayotou, PanosForeign aid has often been envisaged as a tool to boosting growth in aid recipient countries and eradicating poverty that many poor countries on the continent nowadays are confronted with. Whether aid succeeds in fulfilling its aims or not is not easy to answer. The highly controversial issue of aid effectiveness continues to baffle researchers and an ongoing debate around this topic keeps on drawing experts’ attention from all over the world. We have tried to examine aid from both a theoretical and an empirical perspective. The first chapter provides the reader with the necessary background so that he/she gets familiarized with the notion of aid and its purposes. The second chapter examines the welfare and terms of trade analysis of aid, focusing mainly on the so-called “Transfer Paradox” (i.e., the enrichment of the donor country and the impoverishment of the recipient) and the conditions under which it can occur. The third chapter is associated with the relationship between aid and corruption, namely whether aid can potentially foster corruption in recipient countries or not. The next chapter tries to identify the possible reasons for aid giving, namely what factors encourage donors to give aid to recipient countries. The issue of aid effectiveness is examined in the fifth chapter, paying attention to aid’s capability of meeting the Millennium Development Goals (MDGs) by 2015. The last chapter is devoted to the aid-growth nexus that is central in the literature on aid. We try to explain whether aid manages to boost growth in aid recipient countries or not and we illustrate this with two case studies from the real world, one for Bangladesh and one for Uganda.Τεκμήριο Airbnb beyond stars: what reviews reveal(2025-12-18) Kuqja, Franc; Pagratis, Spyros; Dendramis, Yiannis; Dioikitopoulos, EvangelosOnline reviews play a pivotal role in shaping consumer choices on platforms like Airbnb, where textual content offers deeper insights into guest experiences than numerical ratings alone, all while macroeconomic factors like GDP per capita, inflation, and unemployment shape expectations. This study analyzes Airbnb data from eight European cities, first linking pricing to property traits like capacity and ratings, host superhost status, and experience via regressions, then tracing sentiment from review texts over time, spatial city clusters, and economic conditions. NLP converts reviews into sentiment scores, balanced annually per city and aggregated at listing and city-year levels, with sigmoid mapping to Airbnb's 1-5 scale for direct rating comparisons. Token analysis pinpoints words driving positive/negative sentiment, revealing city-specific review tendencies. Key findings: Accommodation capacity emerges as the dominant, consistent price driver, while ratings and host factors show inconsistent impacts. Sentiment uncovers rating inflation—scores cluster at the high end—plus geographic disparities proving uneven guest experiences within cities. Pooled regressions with city fixed effects link economic shifts associatively (not causally) to sentiment fluctuations, though effects remain modest. Ultimately, fusing NLP sentiment analysis, spatial clustering, and macro indicators illuminates platform user dynamics, advancing platform economics and computational social science, despite challenges measuring sentiment in overwhelmingly positive review settings.Τεκμήριο Algorithmic complexity as a barrier to voting manipulation(2023-03-28) Φετάνης, Στυλιανός; Fetanis, Stylianos; Athens University of Economics and Business, Department of Economics; Economides, George; Kammas, Pantelis; Arvanitis, StylianosΑυτή η διατριβή απαντά στο ερώτημα εάν η υπολογιστική πολυπλοκότητα που χαρακτηρίζει οποιονδήποτε αλγόριθμο χειραγώγησης εκλογών θα μπορούσε να αποτελέσει και εμπόδιο μιας τέτοιας απόπειρας. Πρώτον, εντοπίζουμε το κύριο πρόβλημα που προκύπτει στον τομέα της Θεωρίας Κοινωνικής Επιλογής που επιτρέπει τη στρατηγική συμπεριφορά και αλλοιώνει τις προκύπτουσες συναρτήσεις κοινωνικής ευημερίας. Το αποτέλεσμα αυτό επεκτείνεται στη συνέχεια και στις ψηφοφορίες, ως μέσο αποκάλυψης της ιεράρχησης που προκύπτει από μια συνάρτηση κοινωνικής ευημερίας. Δεύτερον, ορίζουμε την απαιτούμενη γλώσσα που θα χρησιμοποιήσουμε για να μελετήσουμε την υπολογιστική πολυπλοκότητα των αλγορίθμων σε όλη την υπόλοιπη διατριβή. Ορίζουμε τις έννοιες της αλγοριθμικής πολυπλοκότητας και των προβλημάτων NPc στην επιστήμη των υπολογιστών. Στη συνέχεια, χρησιμοποιώντας αυτή την γλώσσα, μελετάμε πόσο υπολογιστικά δύσκολη μπορεί να είναι μια χειραγώγηση σε μια ποικιλία κανόνων και συνθηκών ψηφοφορίας (όπως με σταθμισμένες, ή χωρίς σταθμισμένες ψήφους) δείχνοντας ότι, θεωρητικά, η χειραγώγηση μπορεί να είναι "δύσκολη". Ωστόσο, σύντομα συνειδητοποιούμε ότι τέτοια αποτελέσματα βασίζονται κυρίως σε σπάνιες περιπτώσεις και μετατοπίζουμε την προσοχή μας στις περιπτώσεις που είναι πιθανό να συναντήσουμε συχνότερα και μελετάμε πώς μπορούμε να εργαστούμε με κατά προσέγγιση λύσεις. Τέτοιες λύσεις λειτουργούν αρκετά καλά στις περισσότερες περιπτώσεις, γεγονός που δείχνει ότι πρέπει να είμαστε προσεκτικοί όταν χρησιμοποιούμε τα θεωρητικά αποτελέσματα που αναφέρονται παραπάνω. Τέλος, εφαρμόζουμε τις έννοιες και τη γλώσσα που κατασκευάσαμε σε όλη αυτή τη διατριβή στο σύστημα ψηφοφορίας που χρησιμοποιείται στις ελληνικές αυτοδιοικητικές εκλογές (τουλάχιστον μέχρι και το 2019). Μελετάμε πιθανές στρατηγικές χειραγώγησης με βεβαιότητα και υπό αβεβαιότητα.
