MODEL ARMA PADA TREND KUNJUNGAN ORANG INDONESIA KE AUSTRAUA  

Oleh:

Anggraini, Dewi

Jenis: Article from Journal - ilmiah nasional
Dalam koleksi: Kalimantan Scientiae: Sain Dan Teknologi vol. 27 no. 74 (Oct. 2009), page 57-68.
Topik: Arrivals Trend; ARMA Model; Forecasting

Ketersediaan

Perpustakaan PKPM Nomor Panggil: K35 Non-tandon: 1 (dapat dipinjam: 0) Tandon: tidak ada

Lihat Detail Induk

Isi artikel

The arrivals trend of Indonesians to Australia from January 1991 to February 2007 had experienced a slightly pulse trend over the period. This, series had a cycle and seasonal pattern with a peak mostly in every December, and a trough mostly in every May. The pattern represents a gradual increase number of movement at the beginning of the period until reached a peak at 27,900 people in February 1997 with the monthly avetage movement of 7,907 people. Then, it experienced a dramatic decreased in March 1997, recorded by 10,500 people and remained stable in April 1997 to February 2007. In forecasting term a fitted model to data is required. This can be obtained by examining the ACF (Autocorrelation Function)/PACF(Partial Autoconelation Function) of data, diffetenced, and mean-corrected value between the predicted and original data, Autofit modeling approach, Maximum Likelihood Estimator (MLE) method, and tests of randomness on residuals for testing how well the modeI fits the actual series. It is found that the ARMA (Autoregessive Moving Average) (24, 7) is the best model to data with mean-corrected value is 0.08. Due to the relatively small value of mean-corrected, this model is recommended to predict a one year ahead of the data. Therefore,it is forecast that the arrivals trend of Indonesians to Australia, from March 2007 to February 2008, will experience slightly decline with still a trough in May 2007, and a peak on December 2007, predicted by 4,088 and 9,769 people respectively.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali