strapvizpy.display

Module Contents

Functions

plot_ci(sample, rep, bin_size=30, n='auto', ci_level=0.95, ci_random_seed=None, title='', x_axis='Bootstrap Sample Mean', y_axis='Count', path=None)

Makes a histogram of a boostrapped sampling distribution

tabulate_stats(stat, precision=2, estimator=True, alpha=True, path=None)

Makes two tables that summerize the statistics from the bootstrapped
strapvizpy.display.plot_ci(sample, rep, bin_size=30, n='auto', ci_level=0.95, ci_random_seed=None, title='', x_axis='Bootstrap Sample Mean', y_axis='Count', path=None)[source]

Makes a histogram of a boostrapped sampling distribution with its confidence interval and oberserved mean.

Parameters

  • sample (list or numpy.ndarray or pandas.core.series.Series) – sample to bootstrap

  • rep (int) – number of replicates of the distribution

  • int (bin_size =) – a number of bins representing intervals of equal size over the range

  • n (str or int, default="auto") – bootstrap sample size, “auto” specifies using the same size as the sample

  • ci_level (float, default=0.95) – confidence level

  • ci_random_seed (None or int, default=None) – seed for random state

  • title (str, default = "") – title of the histogram

  • x_axis (str, default = "Bootstrap Sample Mean") – name of the x axis

  • y_axis (str, default = "Count") – name of the y axis

  • path (None or str, default = None) – specify the directory to save the figure as .png

Returns

plot – histogram of bootstrap distribution with confidence interval and oberserved mean

Return type

histogram

Examples

>>> plot_ci([1, 2, 3, 4, 5, 6, 7], 1000, n=100, ci_level=0.95,
            ci_random_seed=123, title="Bootstrap")
strapvizpy.display.tabulate_stats(stat, precision=2, estimator=True, alpha=True, path=None)[source]

Makes two tables that summerize the statistics from the bootstrapped samples and the parameters for creating the bootstrapped samples. It also allows you to save the tables in html format.

Parameters

  • stat (dict or tuple) – summary statistics produced by the calculate_boot_stats() function

  • precision (int, default=2) – the precision of the table values

  • estimator (boolean, default=True) – include the bootstrap estimate in the summary statistics table

  • alpha (boolean, default=True) – include the significance level in the summary statistics table

  • path (str, default = None) – specify a path to where the tex files of tables should be saved.

Returns

summary statistics: style object

table summerizing the lower bound and upper bound of the confidence interval,the standard error, the sampling statitic (if estimator = True), and the significance level (if alpha = True). Style objects do not display well in a python shell.

bootstrap parameters: style object

table summerizing the parameters of the bootstrap sampling spficiying the original sample size, number of repititions, the significance level, and the number of samples in each bootstrap if its different from the original sample size. Style objects do not display well in a python shell.

Return type

tuple

Examples

>>> st = calculate_boot_stats([1, 2, 3, 4], 1000, level=0.95, random_seed=123)
>>> stats_table, parameter_table  = tabulate_stats(st)
>>> stats_table
>>> parameter_table