Dissertation Title: MODELING AND FORECASTING THE NUMBER OF TOURISTS IN ALBANIA. A comparison of traditional methods, artificial intelligence and hybrid methods
Authors: Morena BRESHANAJ
-
Institution: University of Tirana, Faculty of Economics, Department of Statistics and Applied Informatics
-
Field of study: Information Systems Economy
-
Publication date: 11.06.2026
-
The dissertation is published in Albanian.
© Copyright: Morena BRESHANAJ
Published by the University of Tirana
Based on legal acts, regulations and policies of the UT
👉 Click here to view the full dissertation (PDF)
Abstract
Tourism demand in Albania has experienced significant growth over the last decade, making its forecasting an essential element for economic planning, policy development and capacity management in the public and private sectors. This thesis deals with the modeling and forecasting of monthly non-resident tourist flows for the period 2016–2024 by integrating traditional statistical methods, contemporary AI methods and hybrid structures. Initially, exploratory analysis highlighted the long-term upward trend and strong summer seasonality, as well as structural disruptions related to the global crisis of 2020–2021. Models such as ARIMA, Theta, Holt–Winters, Linear Regression, SVR, XGBoost, ANN, Prophet, etc. were then tested, using a strict 80/20 train–test split. The results showed that some AI models achieve high accuracy in the metrics. Hybrid methods, which combine the structural robustness of statistical models with the flexibility of algorithmic components, provided the most robust forecasts for the 60-month horizon. Multi-criteria ranking with PROMETHEE II enabled a balanced assessment of the models’ performance. The thesis makes a methodological contribution by proposing a general hybrid framework for medium- and long-term forecasting in tourism, which can also be applied in other regional contexts.
Field: Statistics and applied mathematics in economics
Keywords: Modeling, forecasting, statistical methods, AI methods, hybrid methods, tourism, PROMETHEE II, interactive graphics
Abstract
Tourism demand in Albania has experienced significant growth over the last decade, making its forecasting an essential element for economic planning, policy development, and capacity management in the public and private sectors. This thesis deals with the modeling and forecasting of monthly non-resident tourist flows for the period 2016-2024 by integrating traditional statistical methods, contemporary algorithmic approaches, and hybrid structures. Initially, exploratory analysis highlighted the long-term upward trend and strong summer seasonality, as well as structural disruptions related to the global crisis of 2020-2021. Models such as ARIMA, Theta, Holt–Winters, Linear Regression, SVR, XGBoost, ANN, Prophet, etc. were then tested, using a strict 80/20 train-test split. The results showed that some algorithmic models achieve high accuracy in the metrics. Hybrid methods, which combine the structural robustness of statistical models with the flexibility of algorithmic components, provided the most robust forecasts for the 60-month horizon. Multi-criteria ranking with PROMETHEE II enabled a balanced assessment of the models' performance. The thesis makes a methodological contribution by proposing a general hybrid framework for medium- and long-term forecasting in tourism, which can also be applied in other regional contexts.
Field of study: Statistics and applied mathematics in economics
Keywords: Modeling, forecasting, statistical methods, AI methods, hybrid methods, tourism, PROMETHEE II, interactive graphics