Timezones for PeriodIndex

[changhiskhan]

from SO: http://stackoverflow.com/questions/13030402/pandas-plot-function-ignores-timezone-of-timeseries

[changhiskhan]

xref: #2232

[amelio-vazquez-reina]

I am having the same problem here with the latest stable version of Pandas.

[jreback]

these r not supported

[vgoklani]

same issue, what's not supported?

[jreback]

Periods don't support timezone. Plotting needs to convert to a local tz before creating the periods.

[emunsing]

As a workaround, you can remove the timezone info from the DateTimeIndex of the data which you're wishing to plot using .tz_localize(None) which will preserve the values but strip the timezone information. Matplotlib/pyplot will see these naive datetime values and assume that they are UTC, and will happily plot them with working x-axis ticks.

[jreback]

going to close this issue. This add needless complication because Periods can actually fall in more than 1 tz (e.g. on a DST change), when tz's are quite well supported on DTIs / Timestamps and the conversions are pretty straightforward.

[stroobandt]

This StackExchange response leads to this issue.
Any kind of solution to this issue still seems to be appreciated.

[AvverbioPronome]

@stroobandt this is the solution I ended up using, I chose a timezone for the plot, and go with this:

import matplotlib.pytplot as plt, pandas as pd
df = pd.Dataframe(...)
fig, ax = plt.subplots()
df.plot(ax=ax)
import matplotlib.dates as mdates                                                                                   
from pytz import timezone                                                                                           
ax.xaxis.set_major_formatter(mdates.DateFormatter('%d/%m %H:%M', timezone('Europe/Amsterdam'))) 
fig.show()

[joooeey]

I just tried to reproduce the issue described on StackOverflow and couldn't. Pandas plots the graph in whatever timezone the data is passed. That's exactly what a typical user would expect. The timezone is not mentioned anywhere on the plot.

Code from StackOverflow:

import numpy as np
import matplotlib.pyplot as plt
from pandas import DataFrame, date_range

rng = date_range('1/1/2011', periods=200, freq='S', tz="UTC")
data = DataFrame(np.random.randn(len(rng), 3), index=rng, columns=['A', 'B', 'C'])
data_cet = data.tz_convert("CET")

# plot with data in UTC timezone
fig, ax = plt.subplots()
data[["A", "B"]].plot(ax=ax, grid=True)
plt.show()

# plot with data in CET timezone, but the x-axis remains the same as above
fig, ax = plt.subplots()
data_cet[["A", "B"]].plot(ax=ax, grid=True)
plt.show()

UTC figure:

CET figure:

This is on Pandas 1.2.4.