Rumaidah , Siti (2002) Inferensi mean dari statistik-L dengan metode bootstrap. Undergraduate thesis, FMIPA UNDIP.
PDF Restricted to Repository staff only 2154Kb | ||
| PDF 17Kb | |
| PDF 229Kb | |
| PDF 401Kb | |
| PDF 255Kb | |
| PDF 474Kb | |
PDF Restricted to Repository staff only 704Kb | ||
| PDF 218Kb | |
| PDF 239Kb | |
| PDF 754Kb |
Abstract
Statistik-L Tp = E ciXi,9 merupakan suatu kombinasi tinier dari statistik terurut yang berfungsi sebagai estimator dari suatu parameter. Distribusi statistik-L seringkali tidak diketahui oleh karena itu untuk mengestimasi mean dan variansinya digunakan metode bootstrap yaitu dengan mengestimasi mean, variansi, kovariansi bootstrap dari statistik terurut untuk masing — masing replikasi bootstrap Hari statistik-L. Dengan metode bootstrap didapatkan estimasi variansi mean terpangkas yang lebih kecil daripada estimasi variansi mean sebelum dipangkas. Variansi statistik-L akan semakin kecil dan interval konfidensi mean statistik-L mempunyai selang yang semakin baik dengan meningkatnya jumlah replikasi bootstrap. L-statistics Tu = ciXi:. is a linear combination of order statistics that used to estimate a parameter. Distribution of this L-statistics usually unknown, therefore to estimate mean and its variances are used bootstrap methods by finding the bootstrap mean, variances, covariances of order statistics for each bootstrap replication of L-statistics. With bootstrap methods was obtained variance estimation of trimmed mean that is smaller than variance of mean without trimmed L-statistics variance will be small and confidence interval of its mean to be better as the number of bootstrap replication increase.
Item Type: | Thesis (Undergraduate) |
---|---|
Subjects: | Q Science > QA Mathematics |
Divisions: | Faculty of Science and Mathematics > Department of Mathematics |
ID Code: | 31741 |
Deposited By: | Mr UPT Perpus 1 |
Deposited On: | 24 Nov 2011 13:23 |
Last Modified: | 24 Nov 2011 13:23 |
Repository Staff Only: item control page