Description:
<div style="padding: 0px; margin: 0px; color: transparent; position: absolute; white-space: pre; cursor: text; transform-origin: 0% 0%; left: 94.4px; top: 393.797px; font-size: 20px; font-family: sans-serif; transform: scaleX(0.88915);"><div style="padding: 0px; margin: 0px; position: absolute; cursor: text; transform-origin: 0% 0%; left: 94.4px; top: 393.797px; transform: scaleX(0.88915);"><p class="MsoNormal" style="text-align:justify"><span lang="TR" style="font-size: 10.5pt; line-height: 107%; font-family: "Segoe UI", sans-serif; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;">Tourismdemand is the basis on which all commercial decisions concerning tourismultimately depend. Accurate estimation of tourism demand is essential for thetourism industry because it can help reduce risk and uncertainty as well aseffectively provide basic information for better tourism planning. The purposeof this study is to develop the optimal forecasting model that yields thehighest accuracy when compared to the forecast performances of three differentmethods, namely Artificial Neural Network (ANN), Exponential Smoothing, andBox-Jenkins methods for forecasting monthly inbound tourist flows to Croatia.Prior studies have been applied to forecast tourism demand to Croatia based ontime series models and casual methods. However, the monthly and comparativetourism demand forecasting studies using ANNs are still limited, and this paperaims to fill this gap. The number of monthly foreign tourist arrivals toCroatia covers the period between January 2005-December 2019 data were used tobuild optimal forecasting models. Forecasting performances of the models weremeasured by Mean Absolute Percentage Error (MAPE) statistics. As a result ofthe experiments carried out, when compared to the forecasting performances ofvarious models, 12 lagged ANN models, which have [4-3-1] architecture, wereseen to perform best among all models applied in this study. Considering boththe empirical findings obtained from this study and previous studies on tourismforecasting, it can be seen that ANN models that do not have any negativities(such as over-training, faulty architecture, etc.) produce successfulforecasting results when compared with results generated by conventionalstatistical methods.</span><span lang="TR"><o:p></o:p></span></p></div></div>