What is bootstrapping Be sure to include a discussion of the

What is bootstrapping? Be sure to include a discussion of the general bootstrap algorithm. Choose the best answer below.

Solution

Correct Answer (a).

Bootstrapping is a computer-intensive approach to statistical inference where by parameters are estimated by treating a set of sample data as a population. A computer is used to resample with replacement ‘n’ observations from the sample data. This process is repeated many (say 1000) times. For each resample, the statistic (such as the sample mean) is obtained.

     The bootstrap method was invented by Bradley Efron from Stanford University in 1979.The basic idea behind the bootstrap is ingeniously simple. Take a random sample of ‘n’ observation whose distribution is unknown. Obtain B random samples with replacement and estimate the value of the parameter of interest (say, the mean) for each of the B random samples. Use these B statistics to judge the value of the parameter.

    

      If using the bootstrap method to obtain a confidence interval for a population means. We can use the distribution of the sample means obtained via re-sampling as a proxy for the actual sampling distribution of the sample mean. A (1- ).100% confidence interval could be obtained by determine the /2.100 percentile (lower bound) and the (1- /2).100 percentile (upper bound) of the distribution.