6

If I create a barplot using Seaborn and specify the geometric mean or the median as the estimator, does Seaborn know to use the appropriate standard error formula to create error bars?

Tasos
  • 3,860
  • 4
  • 22
  • 54
LizK
  • 61
  • 2
  • 3
    Seaborn uses [bootstrapped confidence intervals](http://stackoverflow.com/questions/29481134/how-are-the-error-bands-in-seaborn-tsplot-calculated). Alternative estimators are [currently not supported](http://stackoverflow.com/questions/32771520/how-to-use-a-weighted-mean-estimator-in-seaborn-factor-plot-incl-bootstrapping). – Emre Mar 01 '16 at 18:26
  • Does this answer your question? [ValueError: Input contains NaN, infinity or a value too large for dtype('float32')](https://datascience.stackexchange.com/questions/11928/valueerror-input-contains-nan-infinity-or-a-value-too-large-for-dtypefloat32) – mccurcio Aug 17 '20 at 22:44

1 Answers1

1

This information is in the documentation of Seaborn.

They show a bootstrap confidence interval, computed by resampling units (rows in the 2d array input form). By default, in seaborn version 0.8.1 it uses 95% of confidence interval, which is equivalent to a standard error. This value is a parameter that can be changed.

Carlos Mougan
  • 6,011
  • 2
  • 15
  • 45