Just wanna share my experience

Theory of Mean and Variance of Population Estimation Using Jackknife and Bootstrap Method

Theory of Mean and Variance of Population Estimation Using Jackknife and Bootstrap Method

Abstract of my Bachelor Thesis (1999)

Most of statistical methods, such as hypothesis testing, and maximum likelihood estimation, were designed to be implemented on mechanical calculators. These classical methods require distributional assumptions, usually normal distribution. Modern electronic computation has encouraged a development of new statistical method called Resampling that require fewer distributional assumptions than classical methods and can be applied to more complicated statistical estimators.

This research is focusing on two resampling methods, Jackknife and Bootstrap method. The accuracy of the mean and variance estimation is compared by standard error they produce.

The Jackknife method introduced by Quenouille was not only a nonparametric device for estimating bias but also to obtain approximate confidence interval in problems where standard statistical procedures could not be done. One of the promising developments in computer-intensive is Bootstrap method. Bootstrap was introduced primarily as a device for extending the standard error of mean formula to other estimators that the mean. This method can be applied to almost any statistical estimation problems and the data set does not have to be a simple random sample from a single distribution.

Normally distributed data and data with unknown distribution are used. Statistical software (Minitab) and Turbo Pascal are used for resampling calculations.

Bootstrap and Jackknife will have the same accuracy if they are used to estimate mean of population using normally distributed data set. Otherwise, Bootstrap has better accuracy than jackknife for data with unknown distribution. For estimating variance of population in normally distributed data and data with unknown distribution, bootstrap has the highest accuracy. Overall, Bootstrap method is better than Jackknife for estimating mean and variance of population.


May 23, 2008 - Posted by | Statistics

1 Comment »

  1. Salam Tio
    sy mahsiswa tkgt akhir nih,
    tertarik d Bootstrap..

    sy btuh bgt buku berjudul “the jackknife and bootstrap” jun shao
    u punya g?
    please bgt sdh date line nih…
    email sya di march.dian@yahoo.co.id

    Comment by Mardian | December 4, 2008 | Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: