{"cells":[{"cell_type":"markdown","source":["# Time Series\n","\n","```{jupyter-info}\n","{rel-data-download}`bicycles.csv`\n","```\n","\n","We start by loading the biycycles dataset into a `DataFrame` using the same code as in the last slide to index the data by the date."],"metadata":{}},{"cell_type":"code","execution_count":1,"source":["# Some setup code to get the plotting library correct\n","import matplotlib.pyplot as plt\n","import pandas as pd\n","\n","%matplotlib inline"],"outputs":[],"metadata":{}},{"cell_type":"code","execution_count":2,"source":["df = pd.read_csv('bicycles.csv', \n"," index_col='Date', parse_dates=True)\n","\n","# Sorts the rows so the index is sorted\n","df = df.sort_index() \n","\n","df # For display"],"outputs":[{"output_type":"execute_result","data":{"text/plain":[" East West\n","Date \n","2012-10-03 00:00:00 4.0 9.0\n","2012-10-03 01:00:00 4.0 6.0\n","2012-10-03 02:00:00 1.0 1.0\n","2012-10-03 03:00:00 2.0 3.0\n","2012-10-03 04:00:00 6.0 1.0\n","... ... ...\n","2019-03-31 19:00:00 30.0 58.0\n","2019-03-31 20:00:00 26.0 31.0\n","2019-03-31 21:00:00 18.0 15.0\n","2019-03-31 22:00:00 7.0 14.0\n","2019-03-31 23:00:00 6.0 10.0\n","\n","[56904 rows x 2 columns]"],"text/html":["
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
EastWest
Date
2012-10-03 00:00:004.09.0
2012-10-03 01:00:004.06.0
2012-10-03 02:00:001.01.0
2012-10-03 03:00:002.03.0
2012-10-03 04:00:006.01.0
.........
2019-03-31 19:00:0030.058.0
2019-03-31 20:00:0026.031.0
2019-03-31 21:00:0018.015.0
2019-03-31 22:00:007.014.0
2019-03-31 23:00:006.010.0
\n","

56904 rows × 2 columns

\n","
"]},"metadata":{},"execution_count":2}],"metadata":{}},{"cell_type":"markdown","source":["Just like with any other `pandas` object, we can inspect the `index` of this `DataFrame` using the `.index` attribute."],"metadata":{}},{"cell_type":"code","execution_count":3,"source":["df.index"],"outputs":[{"output_type":"execute_result","data":{"text/plain":["DatetimeIndex(['2012-10-03 00:00:00', '2012-10-03 01:00:00',\n"," '2012-10-03 02:00:00', '2012-10-03 03:00:00',\n"," '2012-10-03 04:00:00', '2012-10-03 05:00:00',\n"," '2012-10-03 06:00:00', '2012-10-03 07:00:00',\n"," '2012-10-03 08:00:00', '2012-10-03 09:00:00',\n"," ...\n"," '2019-03-31 14:00:00', '2019-03-31 15:00:00',\n"," '2019-03-31 16:00:00', '2019-03-31 17:00:00',\n"," '2019-03-31 18:00:00', '2019-03-31 19:00:00',\n"," '2019-03-31 20:00:00', '2019-03-31 21:00:00',\n"," '2019-03-31 22:00:00', '2019-03-31 23:00:00'],\n"," dtype='datetime64[ns]', name='Date', length=56904, freq=None)"]},"metadata":{},"execution_count":3}],"metadata":{}},{"cell_type":"markdown","source":["So now to get a row for a particular date and time, we can index into the `DataFrame` using `loc`! The key difference here is that we will specify a string for the date-time rather than a number (since the index is the date-time)."],"metadata":{}},{"cell_type":"code","execution_count":4,"source":["df.loc['2019-03-31 15:00:00']"],"outputs":[{"output_type":"execute_result","data":{"text/plain":["East 130.0\n","West 121.0\n","Name: 2019-03-31 15:00:00, dtype: float64"]},"metadata":{},"execution_count":4}],"metadata":{}},{"cell_type":"markdown","source":["The incredibly powerful thing about using date-time as the index type is it allows us to do semantic indexing based on dates and times. Here are some examples:\n","\n","You can pick just a date and it will include all rows from that date (meaning it will have all the times for that day)."],"metadata":{}},{"cell_type":"code","execution_count":5,"source":["df.loc['2017-03-31']"],"outputs":[{"output_type":"execute_result","data":{"text/plain":[" East West\n","Date \n","2017-03-31 00:00:00 2.0 4.0\n","2017-03-31 01:00:00 2.0 0.0\n","2017-03-31 02:00:00 1.0 0.0\n","2017-03-31 03:00:00 1.0 0.0\n","2017-03-31 04:00:00 3.0 3.0\n","2017-03-31 05:00:00 23.0 14.0\n","2017-03-31 06:00:00 57.0 49.0\n","2017-03-31 07:00:00 163.0 99.0\n","2017-03-31 08:00:00 250.0 162.0\n","2017-03-31 09:00:00 90.0 70.0\n","2017-03-31 10:00:00 52.0 38.0\n","2017-03-31 11:00:00 39.0 31.0\n","2017-03-31 12:00:00 37.0 37.0\n","2017-03-31 13:00:00 52.0 34.0\n","2017-03-31 14:00:00 45.0 55.0\n","2017-03-31 15:00:00 74.0 87.0\n","2017-03-31 16:00:00 83.0 223.0\n","2017-03-31 17:00:00 145.0 333.0\n","2017-03-31 18:00:00 117.0 206.0\n","2017-03-31 19:00:00 50.0 84.0\n","2017-03-31 20:00:00 26.0 36.0\n","2017-03-31 21:00:00 16.0 40.0\n","2017-03-31 22:00:00 12.0 22.0\n","2017-03-31 23:00:00 5.0 15.0"],"text/html":["
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
EastWest
Date
2017-03-31 00:00:002.04.0
2017-03-31 01:00:002.00.0
2017-03-31 02:00:001.00.0
2017-03-31 03:00:001.00.0
2017-03-31 04:00:003.03.0
2017-03-31 05:00:0023.014.0
2017-03-31 06:00:0057.049.0
2017-03-31 07:00:00163.099.0
2017-03-31 08:00:00250.0162.0
2017-03-31 09:00:0090.070.0
2017-03-31 10:00:0052.038.0
2017-03-31 11:00:0039.031.0
2017-03-31 12:00:0037.037.0
2017-03-31 13:00:0052.034.0
2017-03-31 14:00:0045.055.0
2017-03-31 15:00:0074.087.0
2017-03-31 16:00:0083.0223.0
2017-03-31 17:00:00145.0333.0
2017-03-31 18:00:00117.0206.0
2017-03-31 19:00:0050.084.0
2017-03-31 20:00:0026.036.0
2017-03-31 21:00:0016.040.0
2017-03-31 22:00:0012.022.0
2017-03-31 23:00:005.015.0
\n","
"]},"metadata":{},"execution_count":5}],"metadata":{}},{"cell_type":"markdown","source":["You could also leave off the day to get all of the rows for a year and month."],"metadata":{}},{"cell_type":"code","execution_count":6,"source":["df.loc['2017-03']"],"outputs":[{"output_type":"execute_result","data":{"text/plain":[" East West\n","Date \n","2017-03-01 00:00:00 1.0 2.0\n","2017-03-01 01:00:00 2.0 2.0\n","2017-03-01 02:00:00 1.0 1.0\n","2017-03-01 03:00:00 1.0 0.0\n","2017-03-01 04:00:00 3.0 3.0\n","... ... ...\n","2017-03-31 19:00:00 50.0 84.0\n","2017-03-31 20:00:00 26.0 36.0\n","2017-03-31 21:00:00 16.0 40.0\n","2017-03-31 22:00:00 12.0 22.0\n","2017-03-31 23:00:00 5.0 15.0\n","\n","[744 rows x 2 columns]"],"text/html":["
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
EastWest
Date
2017-03-01 00:00:001.02.0
2017-03-01 01:00:002.02.0
2017-03-01 02:00:001.01.0
2017-03-01 03:00:001.00.0
2017-03-01 04:00:003.03.0
.........
2017-03-31 19:00:0050.084.0
2017-03-31 20:00:0026.036.0
2017-03-31 21:00:0016.040.0
2017-03-31 22:00:0012.022.0
2017-03-31 23:00:005.015.0
\n","

744 rows × 2 columns

\n","
"]},"metadata":{},"execution_count":6}],"metadata":{}},{"cell_type":"markdown","source":["Unsurprisingly, you can also just get all the rows for a year. All of these examples are accomplished by interpreting the value you are using to index as a date-time, and then selecting all the rows that match that date-time. If you just specify a year, it finds all rows that have that year."],"metadata":{}},{"cell_type":"code","execution_count":7,"source":["df.loc['2017']"],"outputs":[{"output_type":"execute_result","data":{"text/plain":[" East West\n","Date \n","2017-01-01 00:00:00 0.0 5.0\n","2017-01-01 01:00:00 5.0 14.0\n","2017-01-01 02:00:00 1.0 0.0\n","2017-01-01 03:00:00 0.0 2.0\n","2017-01-01 04:00:00 0.0 1.0\n","... ... ...\n","2017-12-31 19:00:00 9.0 12.0\n","2017-12-31 20:00:00 6.0 8.0\n","2017-12-31 21:00:00 3.0 10.0\n","2017-12-31 22:00:00 7.0 6.0\n","2017-12-31 23:00:00 7.0 9.0\n","\n","[8760 rows x 2 columns]"],"text/html":["
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
EastWest
Date
2017-01-01 00:00:000.05.0
2017-01-01 01:00:005.014.0
2017-01-01 02:00:001.00.0
2017-01-01 03:00:000.02.0
2017-01-01 04:00:000.01.0
.........
2017-12-31 19:00:009.012.0
2017-12-31 20:00:006.08.0
2017-12-31 21:00:003.010.0
2017-12-31 22:00:007.06.0
2017-12-31 23:00:007.09.0
\n","

8760 rows × 2 columns

\n","
"]},"metadata":{},"execution_count":7}],"metadata":{}},{"cell_type":"markdown","source":["You can also use ranges to select multiple years!"],"metadata":{}},{"cell_type":"code","execution_count":8,"source":["df.loc['2017':'2018'] # All rows from 2017 to 2018 (inclusive for pandas)"],"outputs":[{"output_type":"execute_result","data":{"text/plain":[" East West\n","Date \n","2017-01-01 00:00:00 0.0 5.0\n","2017-01-01 01:00:00 5.0 14.0\n","2017-01-01 02:00:00 1.0 0.0\n","2017-01-01 03:00:00 0.0 2.0\n","2017-01-01 04:00:00 0.0 1.0\n","... ... ...\n","2018-12-31 19:00:00 9.0 5.0\n","2018-12-31 20:00:00 12.0 14.0\n","2018-12-31 21:00:00 7.0 7.0\n","2018-12-31 22:00:00 3.0 4.0\n","2018-12-31 23:00:00 7.0 6.0\n","\n","[17520 rows x 2 columns]"],"text/html":["
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
EastWest
Date
2017-01-01 00:00:000.05.0
2017-01-01 01:00:005.014.0
2017-01-01 02:00:001.00.0
2017-01-01 03:00:000.02.0
2017-01-01 04:00:000.01.0
.........
2018-12-31 19:00:009.05.0
2018-12-31 20:00:0012.014.0
2018-12-31 21:00:007.07.0
2018-12-31 22:00:003.04.0
2018-12-31 23:00:007.06.0
\n","

17520 rows × 2 columns

\n","
"]},"metadata":{},"execution_count":8}],"metadata":{}},{"cell_type":"markdown","source":["# Plotting\n","In Lesson 10, we will spend time discussing how to plot data and how to interpret those visualizations. `pandas` provides some basic functionality for plotting."],"metadata":{}},{"cell_type":"code","execution_count":9,"source":["df.plot()"],"outputs":[{"output_type":"execute_result","data":{"text/plain":[""]},"metadata":{},"execution_count":9},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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"},"metadata":{"needs_background":"light"}}],"metadata":{}},{"cell_type":"markdown","source":["This is not exactly readable... The problem is the data is too \"high resolution\": it's drawing a line for every hour of every day for 6 years! No wonder we can't actually see anything.\n","\n","What we need is to \"resample\" the data so the data points occurr less frequently. One way to do this would be to just drop every row besides Sundays at 12pm (arbitrarily decided), but you can tell that won't work because it won't give us a good idea of overall biking trends. Instead, it would be nice if we could re-organize the data so each row was the total number of bikers in a week (just for example).\n","\n","This sounds somewhat similar to a group-by, but turns out to work quite differently (more on this in next section). To change the time series to have the granularity of every week and then plot that, we use the `resample` method. In the next cell, we transform the data so each row is a time-span of a week and has the sum of all the bikers for the hours in that week."],"metadata":{}},{"cell_type":"code","execution_count":10,"source":["weekly = df.resample('W').sum()\n","weekly.plot()"],"outputs":[{"output_type":"execute_result","data":{"text/plain":[""]},"metadata":{},"execution_count":10},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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"},"metadata":{"needs_background":"light"}}],"metadata":{}},{"cell_type":"markdown","source":["Much more legible! Now we can see both East and West traffic. Looking at the graphs, we see some seasonality in the data - there are peaks in the summer and troughs in the winter. \n","\n","`'W'` is a special code for `resample` to tell it to resample by week. There are many other codes you can pass to it instead!\n","* `'D'` = day\n","* `'W'` = week\n","* `'M'` = month\n","* `'A'` = year\n","* [And much much more!](https://pandas.pydata.org/pandas-docs/version/0.17.0/timeseries.html#offset-aliases)\n","\n","Generally when analyzing time series, you are interested in looking at two metrics of describing the series.\n","* Its **trend**, which shows how the average value changes over time. This dataset seems to have no obvious upward or downward trend (the average number of bicyclers, in say a year, are about the same).\n","* Its **seasonality**, which shows how the sequence fluctuates over some period. This series definitely has some seasonality to it since it goes up and down in relatively the same pattern each year.\n","\n","# Resample vs Group By\n","Earlier, we showed that to change the time series to be by week, instead of by hour, we used a `resample`. We commented that we didn't actually want to use `groupby` because these are fundamentally different operations.\n","\n","To show you that these in-fact, are different, we can try using a `groupby` to solve the problem. If we want to access the `week` for each row, we can use `df.index.week`. `df.index` returns the date-time index and `.week` gets the week value from each row. The week value will be a number from 0 to 52, indicating which week that date-time is in.\n","\n","In the cell below, we `groupby` this week value and then plot the result."],"metadata":{}},{"cell_type":"code","execution_count":11,"source":["weekly_groupby = df.groupby(df.index.week).sum()\n","weekly_groupby.plot()"],"outputs":[{"output_type":"execute_result","data":{"text/plain":[""]},"metadata":{},"execution_count":11},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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"},"metadata":{"needs_background":"light"}}],"metadata":{}},{"cell_type":"markdown","source":["Wow! Why does that look so different?!?!\n","\n","It has to do with how `groupby` works and the fact that we tried to group by the week-number. As we said before, `df.index.week` will be a `Series` of numbers between 0 and 52, indicating which week the date-time fell in. However, this causes a problem for the same week in different years! \n","\n","Consider `Jan 2, 2018` and `Jan 5, 2019`. Both of those `.week` values will be `0` since they are in the first week! The year information was lost! They will end up landing in the same group, which is why we can see that graph only goes from 0 to 52!\n","\n","Instead, we want to do some kind of grouping based on the weeks themselves. We want `Jan 2, 2018` and `Jan 5, 2019` to go into separate groups with their respective date-times that are in the same week. \n","\n","While it is actually possible to try to put something together that works and uses `groupby`, we will not go into that here. Instead, `resample` is an operation precisely meant to make this easier! You just tell `pandas` you want to resample by week, and it figures that out for you! \n","\n","There is a lot more complexity you can get into with time series. For example, `resample` turns out to be much more complicated since you can use it to **downsample** (throw away data) or **upsample** (create new data) so that the data are sampled at the desired frequency. There are too many details to go into that here, but you may learn more from [`pandas` documentation](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.resample.html)!"],"metadata":{}}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.8.3"}},"nbformat":4,"nbformat_minor":2}