Aileen Nielsen - Time Series Analysis - PyCon 2017

"Speaker: Aileen Nielsen
Time series analysis is more relevant than ever with the rise of big data, the internet of things, and the general availability of data that follows events through time. This tutorial will introduce participants to the many versatile tools Python offers for exploring, analyzing, and predicting time series data. The tutorial will be a mix of lecture and practice, and it will be broken down into four components:
(1) Handling timestamped data in Python
(2) Commonly encountered problems with time series
(3) Time series prediction exercises
(4) Time series classification exercises
Slides can be found at: speakerdeck.com/pycon2017 and github.com/PyCon/2017-slides"

[@wenzhongtan1358]

1:13:30 Moving window function explanation 1:20:50 Moving window function code 1:50:25 Self-correlation explanation 1:57:45 Self-correlation code 2:05:55 Munging and stationarity explanation 2:15:00 Forecasting explanation 2:24:30 AR+MA processes code 2:47:55 Clustering & classification explanation 2:53:05 Clustering & classification code 3:07:27 Other explanations: Vector autoregression, Kalman filter, hidden Markov models, recurrent neural networks

[@TheEntireUniverse]

52:00 if you are already familiar with pandas and pandas date/time functionality.

[@Ammar_Abdilghanie]

great talk Aileen! thank you for sharing your knowledge! you are an excellent teacher, totally enjoyed your presentation

[@adithyapai6158]

Thanks a lot...gives sense of direction to understand the concept. But for intermediate level.

[@HarishNarayanan]

Thank you so much for this, it was really informative.

[@wasgeht2409]

OMG , the best Video about Time Series

[@ploxedc]

This is the most epic video of all time

[@mirandamoralez6354]

Thanks a lot, very nice job!

[@mirandamoralez6354]

ps.: for me, best explanation of acf and pacf - and i ve seen a lot of videos and papers about this topic so far!!

[@yaswanthkumargothireddy]

+PyCon 2017 please update the notebook to github. Old one was not organized and I am unable to follow with the video.

[@tobspatterson]

I'm not sure about the origin of this repo, but checkout github.com/ikding/pycon_time_series .

[@aminemehablia18]

Thank you for the talk, really appreciated. Where to find the code used ?

[@sathyanyn]

12.5 - TimeSeries Begining

[@vivekpuurkayastha1580]

What is the difference between the 2016 talk and this one?

[@lesternygaard842]

59:06 no that's because you wrote `converted[1:10]` instead of `[0:10]` (or `[:10]`)

[@rohullahalavi]

like

[@browniris6289]

she did such a bad job of explaining when to use rolling functions over extending functions! confused me on what i knew already :/

[@Eizengoldt]

Make a better one yourself then

[@yashsharma682]

The presenter is really weird

[@ashugurtoo4835]

When I was doing PhD in physics ... How old is this chick.... Good vedio though..

[@yasararafath535]

She was a lawyer, before she dwelled into physics. May be that's why she's presenting in SciPy, Pycon annual conventions. Not sure about PhD though.