Atmanegara, Eviyana (2019) Peramalan Kunjungan Wisatawan Mancanegara ke Indonesia dengan Analisis Intervensi pada Count Time Series. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
0621175001702 Master_Thesis.pdf - Accepted Version Download (3MB) | Preview |
Abstract
Data jumlah kunjungan wisatawan mancanegara adalah data count time series yang berisi nilai diskrit. Poisson autoregressive (Poisson-AR) dan Binomial Negatif autoregressive (BN-AR) adalah model time series yang digunakan untuk meramalkan data count. Jumlah kedatangan wisatawan mancanegara dipengaruhi oleh serangkaian input yang disebut intervensi, seperti kejadian teror bom dan promosi pariwisata. Penelitian ini bertujuan untuk memperkirakan jumlah wisatawan mancanegara yang berkunjung ke Indonesia berdasarkan kebangsaan. Jumlah kedatangan wisatawan dari Bahrain dan Singapura masing-masing mewakili data jumlah sedikit dan data jumlah banyak. Pemodelan dilakukan dengan model intervensi Poisson-AR, BN-AR, dan Gaussian-AR. Evaluasi kebaikan model dilakukan dengan membandingkan nilai RMSE (Root Mean Square Error), MAD (Mean absolute deviation), MAPE (Mean Absolute Percentage Error) pada data out-of-sample. Model dengan nilai RMSE, MAD, dan MAPE terkecil merupakan model terbaik. Hasil penelitian menunjukkan bahwa model intervensi Poisson-AR merupakan model terbaik pada peramalan data out-of-sample kunjungan wisman berkewarganegaraan Singapura dan bahrain. Hasil ramalan menunjukkan peningkatan jumlah wisman berkewarganegaraan Singapura. Pada Tahun 2019 diramalkan jumlah kunjungan wisman berkewarganegaraan Singapura dan Bahrain berturut-turut 2170253 dan 1639 kunjungan.===================================================================================The foreign tourist's data are count time series data that contains the discrete value. Poisson autoregressive (Poisson-AR) and Negative Binomial autoregressive (NB-AR) are time series models used for forecasting count data. The number of foreign tourist arrivals is influenced by the series of inputs called interventions, such as the existence of bomb terror and tourism promotion. This research aims to forecast the number of foreign tourists visiting Indonesia by nationality. The number of tourist arrivals from Bahrain and Singapore represents low count data and high count data, respectively. This work employs intervention on Poisson-AR, NB-AR, and Gaussian-AR model. Evaluation of the model's goodness is done by comparing the value of RMSE (Root Mean Square Error), MAD (Mean absolute deviation), MAPE (Mean Absolute Percentage Error). The results showed that the intervention model on Poisson AR was the best. The model is used to estimate the number of tourist arrivals from Singapore and Bahrain to Indonesia. The forecast results show an increase in the number of tourists from Singapore. In 2019 the number of foreign tourists from Singapore and Bahrain is predicted to be 2170253 and 1639 visits respectively.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | BN-AR, Count Time Series, Intervensi, Poisson-AR, Wisatawan Mancanegara. |
| Subjects: | Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Mathematics > 44101-(S2) Master Thesis |
| Depositing User: | Eviyana Atmanegara |
| Date Deposited: | 20 May 2021 05:46 |
| Last Modified: | 20 May 2021 05:46 |
| URI: | http://repository.its.ac.id/id/eprint/68222 |
Actions (login required)
 |
View Item |