10 Dates and Times in `pandas`

For these examples, we will use a time-series dataset of stock prices:

Code

from pandas import read_csv

request_url = "https://raw.githubusercontent.com/prof-rossetti/intro-to-python/main/data/daily_adjusted_nflx.csv"
df = read_csv(request_url)
df.drop(columns=["split_coefficient", "dividend_amount"], inplace=True)
df.head()

	timestamp	open	high	low	close	adjusted_close	volume
0	2021-10-18	632.1000	638.4100	620.5901	637.97	637.97	4669071
1	2021-10-15	638.0000	639.4200	625.1600	628.29	628.29	4116874
2	2021-10-14	632.2300	636.8800	626.7900	633.80	633.80	2672535
3	2021-10-13	632.1791	632.1791	622.1000	629.76	629.76	2424638
4	2021-10-12	633.0200	637.6550	621.9900	624.94	624.94	3227349

10.1 Converting to Datetime

When we are working with a column of strings representing date and/or time values, it is helpful to use the to_datetime function to convert these string values into datetime-aware values that respect datetime operations.

Here we see the original values in the “timestamp” column are date strings:

type(df["timestamp"][0])

str

After converting this column of string values to datetime-aware values, by passing that column to the to_datetime function, we see the resulting values become datetime-aware Timestamp objects:

from pandas import to_datetime

df["timestamp"] = to_datetime(df["timestamp"])

type(df["timestamp"][0])

pandas._libs.tslibs.timestamps.Timestamp

10.2 Datetime Operations

When working with Timestamp objects, we can easily convert and transform them to various component date parts (such as just the year, just the month, etc.):

df["year"] = df["timestamp"].dt.year
df["month"] = df["timestamp"].dt.month
df["weekday"] = df["timestamp"].dt.weekday
df["day_name"] = df["timestamp"].dt.strftime("%A") # DAY NAME

df[["timestamp", "year", "month", "weekday", "day_name"]]

	timestamp	year	month	weekday	day_name
0	2021-10-18	2021	10	0	Monday
1	2021-10-15	2021	10	4	Friday
2	2021-10-14	2021	10	3	Thursday
3	2021-10-13	2021	10	2	Wednesday
4	2021-10-12	2021	10	1	Tuesday
...	...	...	...	...	...
95	2021-06-03	2021	6	3	Thursday
96	2021-06-02	2021	6	2	Wednesday
97	2021-06-01	2021	6	1	Tuesday
98	2021-05-28	2021	5	4	Friday
99	2021-05-27	2021	5	3	Thursday

100 rows × 5 columns

For more information about date and time conversions, see the following documentation links: