Naslov
Artificial neural network analysis of domestic tourism in Croatia

Autori
Gregorić, Maja

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Economic and social development (Book of proceedings), 76th International scientific conference on economic and social development development - Building resilient society / Misevic, Petar ; Kontic, Ljiljana ; Galovic, Tomislav - Zagreb : VADEA ; Sveučilište Sjever, 2021, 47-55

Skup
76th International Scientific Conference on Economic and Social Development: "Building Resilient Society"

Mjesto i datum
Zagreb, Hrvatska, 17.12.2021. - 18.12.2021

Vrsta sudjelovanja
Ostalo

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
domestic tourists ; artificial neural network ; forecasting

Sažetak
The purpose of this paper is to investigate which variables have an impact on the arrivals and overnight stays of domestic tourists. The research question is how to effectively model time series that exhibit seasonal patterns. The number of tourist arrivals and overnight stays of domestic tourists from January 2005 to March 2020 is used to model an artificial neural network. A dataset with the following independent variables was attempted as model input: Consumer Price Index (previous month=100), Average Monthly Net Salary at Nominal Prices, Consumer Confidence Index, Consumer Sentiment Index and Consumer Expectation Index. Data were seasonally adjusted using the X- 12 ARIMA seasonal adjustment procedure. The best- fitting model is the one that achieved a mean absolute percentage error of 5.32%, which represents a high forecasting accuracy that is essential for further activities in the tourism sector and important for all those involved in the tourism process.

Izvorni jezik
Engleski

Znanstvena područja
Ekonomija

POVEZANOST RADA