Rahmawati, Eriza (2014) Kajian Ukuran Sampel Bootstrap Pada Pemodelan Struktural Teknik Partial Least Square (PLS). Sarjana thesis, Universitas Brawijaya.
Abstract
Penelitian dengan sampel kecil seringkali berasal dari populasi yang tidak normal. Untuk mengatasi masalah tersebut dapat digunakan metode resampling seperti Bootstrap. Salah satu teknik pemodelan struktural yang mengggunakan metode resampling adalah Partial Least Square (PLS). Penelitian ini mengkaji mengenai ukuran sampel Bootstrap yaitu B dan n elemen Bootstrap untuk mengetahui apakah terdapat perbedaan persentase signifikansi dan pada ukuran sampel ke berapa persentase signifikansi konvergen. Dari 5 data yang dianalisis, 3 data dengan ukuran sampel asli 80-100 tidak menunjukkan adanya perbedaan persentase signifikansi, sedangkan 2 data dengan ukuran sampel asli sebesar 50-55 menunjukkan adanya perubahan persentase signifikansi. Persentase signifikansi konvergen ketika ukuran sampel Bootstrap B=100 dengan n elemen sebesar sampel asli atau maksimal berkurang 10% dari sampel asli, dengan ukuran sampel asli minimal 50-55.
English Abstract
Research with a small sample size is often derived from a population that is not normal. To overcome it, can be used as Bootstrap resampling method. One of the structural modeling techniques that use resampling method is Partial Least Square (PLS). This research was examines the Bootstrap sample size that is B and n elements to determine whether there are differences in the percentage of significance and to what sample size the percentage of the significance convergent. From the 5 data were analyzed, the 3 data of which with the original sample size of 80-100 showed no difference in the percentage of significance, while the 2 other data with the original sample size of 50-55 shows a chance in the percentage of significance. Percentage of significance convergent when the sample size Bootstrap is B = 100 and the n elements is the same with the original sample size or maximum 10% reduction from the original sample, with the original sample size at least 50-55.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/MIPA/2014/171/051403714 |
| Subjects: | 500 Natural sciences and mathematics > 519 Probabilities and applied mathematics > 519.5 Statistical mathematics |
| Divisions: | Fakultas Matematika dan Ilmu Pengetahuan Alam > Statistika |
| Depositing User: | Hasbi |
| Date Deposited: | 08 Jul 2014 09:19 |
| Last Modified: | 09 Nov 2021 07:22 |
| URI: | http://repository.ub.ac.id/id/eprint/153789 |
Preview |
Text
SKRIPSI_DAFTAR_PUSTAKA.pdf Download (1MB) | Preview |
Preview |
Text
SKRIPSI_LAMPIRAN.pdf Download (2MB) | Preview |
Preview |
Text
SKRIPSI_COVER-ABSTRAK-DAFTAR_ISI.pdf Download (1MB) | Preview |
Preview |
Text
SKRIPSI_GABUNG.pdf Download (2MB) | Preview |
Actions (login required)
 |
View Item |