Consider the examples for tsplot (time-series plot) in seaborn, e.g. the one shown below.
Do pandas, matplotlib or seaborn include similar functionality to plot translucent bands representing percentiles or standard deviations of a population provided in a dataframe?
Reading the documentation, it looks like sns.tsplot plots confidence intervals for the sampling distribution of the estimator via bootstrapping, which is of course different.
Worth mentioning that the proposal by raphael conflates percentiles and confidence intervals, which are very different statistics. For example, if you use ci=[0,100] in tsplot, you will not get bands showing the minimum and maximum of your data. An issue requesting inclusion of percentiles was raised on the Seaborn github page in January 2016, with a clear--and curt--reply that this was not currently possible.
A potential work-around has been posted here.
From the seaborn tsplot documentation, it seems you can provide multiple percentile confidence bands. The example below plots the 68th percentile and 95th percentile confidence bands (ci=[68, 95]). You can also change the value of the line to, for example, the median with estimator=np.median or np.mean.
import numpy as np; np.random.seed(22)
import seaborn as sns; sns.set(color_codes=True)
x = np.linspace(0, 15, 31)
data = np.sin(x) + np.random.rand(10, 31) + np.random.randn(10, 1)
ax = sns.tsplot(data=data, ci=[68, 95], color="m")
