Using StatCrunch to Bootstrap
Step 1: Enter the sample into the StatCrunch spreadsheet.
Step 2: Select Data > Sample. Choose the column. The sample size equals the number of observations in
the data set. Choose at least 5000 samples. Check the “Sample with replacement” box. Check the “Stack
ed with a sample id” radio button. Enter the seed. Check the box “Open in a new data table”.
Step 3: Compute the mean for each bootstrap sample: Stat > Summary Stats > Columns. Choose the
column that contains the bootstrap sample data. Under “Group by:” select “Sample”. Under Statistics,
select “Mean”. Check the box “Store in data table”. Click Compute!.
Step 4: Compute the 2.5
th
and 97.5
th
percentiles (for a 95% confidence interval). Stat > Summary Stats >
Columns. Choose the column that contains the bootstrap means. Enter 2.5,97.5 in the “Percentiles” box.
Click Compute!.
Using the StatCrunch Bootstrap Applet
Step 1: Select Applets > Resampling > Bootstrap a statistic.
Step 2: Choose the data under “Samples in:” Choose the statistic under “Statistic:”. Click Compute!
Step 3: Click “1000 times” at least five times.
Using Minitab to Bootstrap
Step 1: Enter the raw data into column C1.
Step 2: Select the Calc menu, highlight Random Data , then highlight Sample from Columns N . Multiply
the number of observations in the data set by B , the number of resamples (usually 5000) and enter this
result in the cell marked “Number of rows to sample:” For example, if there are 12 observations in the
data set, multiply 12 by 5000 and enter 60,000 in the cell. In the “From columns:” cell, enter C1, the
column containing the raw data. Enter C2 in the “Store samples in:” cell. Check “Sample with
replacement” and hit OK.
Step 3: Select the Calc menu, highlight Make Patterned Data , then highlight Simple Set of Numbers N .
Enter C3 in the cell labeled “Store patterned data in:” Enter 1 in the cell “From first value:” , enter 5000
in the cell “To last value:” , enter 1 in the cell “In steps of:.” In the cell “Number of times to list each
value:” enter the number of observations in the original data set. If the original data set has 12
observations, enter 12. Enter 1 in the cell “Number of times to list the sequence:” Press OK.
Step 4: Select the Stat menu, highlight Basic Statistics , then highlight Store Descriptive Statistics N . In
the cell “Variables:” enter C2, in the cell “By variables:” enter C3. Press the Statistics cbutton and check
the statistic whose value you are estimating. Click OK twice.
Step 5: Now identify the percentiles that represent the middle
(
1 − 𝛼
)
∙ 100% values of the statistic. For
example, to find a 95% confidence interval, identify the 2.5
th
percentile and the 97.5
th
percentile. These
values represent the lower and upper bound of the confidence interval. To find percentiles, select Calc,
then Calculator… . In the Expression box type: Percentile(data, percentile). For example, if you have a
column tilted “Bootstrap Means” and want the 2.5
th
percentile, type Percentile(‘Bootstrap Means’, 0.025).
Enter the column you want the result reported in the “Store result in” cell.