Perbandingan Analisis Trend dan Holt Double Eksponensial Smoothing dalam Meramalkan Angka Kematian Bayi di Jawa Timur  

Oleh:

Qoyyimah, Mazro'atul ; Salim, Lutfi Agus

Jenis: Article from Journal - ilmiah nasional
Dalam koleksi: The Indonesian Journal of Public Health vol. 03 no. 03 (Mar. 2007), page 108.
Topik: trend analysis; Holt double exponensial smoothing; forecasting; infant mortality rate

Ketersediaan

Perpustakaan FK Nomor Panggil: I10.K.01 Non-tandon: 1 (dapat dipinjam: 0) Tandon: tidak ada

Lihat Detail Induk

Isi artikel

forecasting is needed to determine when an event will happen so the exact decision can be implemented. Time series analysis is an exact approach for forecasting a variable in the future, and it will be implemented for the infant mortality data in East Java. The tred analysis and holt double exponential smoothing are times series methods that be used to forecast infant mortality rest in East Java in the future systematically and statistically. The objective of this research was to compare the forecasting result of both merhods to select which method can forecast infant mortality rate in Easst Java accurately according to the smallest MAPE. This research used secondary data of infant mortality rate in East Java started at 1995 until 2005, which were taken from national survey of national survey for social economic (susenas) data by center of statistical board (BPS) of east java. the data were analyzed by using trend analysis, holt double exponential smoothing, and computer program. According to the result of research, the trend analysis (linear) method had found MAPE 3,88% and the MAPE of holt double exponential smoothing methods 4,96%, so the conclusion of this research found the best method of forecasting infant mortality rate in East Java was trend analysis (linear) method with the smallest MAPE. The result hopped can be useful as idea and early information for the planning of health care program in order to decrease infant mortality rate in East Java. To have accurate result should evaluate the re-prediction in short time.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali