CODEWITHSUNDEEP
pandasmoving-average
D.Banerjee
D.Banerjee

Reputation: 91

Calculating rolling mean of 15 minutes, 8 hours and 24 hours using Python

I have a dataframe which contains data for every 10 seconds over a period of one month. I need to calculate the moving average of the data for 15 minutes, 8 hours and 24 hours. How can I do it using Python?

Here is my dataset:

RecTime NO2_RAW NO2 Ox_RAW  Ox  CO_RAW  CO  SO2_RAW SO2
05/31/19 13:42  0   19.13   2225    4480.35 6503.1  7687.33 -3183.6 -8181.55
05/31/19 13:42  17  51.64   1711.2  3454.96 6502.7  7686.86 -3183.6 -8181.55
05/31/19 13:42  48.4    111.69  1387.7  2809.35 6501.9  7685.93 -3183.6 -8181.55
05/31/19 13:42  60.1    134.07  1173.6  2382.07 6501.4  7685.35 -3183.6 -8181.55
05/31/19 13:42  63.9    141.33  1025.6  2086.7  6501    7684.88 -3183.6 -8181.55

I have tried using the following code

> Gas_432_15min = Gas_432.resample(rule='15Min', on='RecTime').mean()

But I think it is not calculating a rolling mean.

Upvotes: 1

Views: 1626

Answers (1)

Quang Hoang
Quang Hoang

Reputation: 150785

There's rolling that does what you want, but require datetimeIndex:

Gas_432.RecTime = pd.to_datetime(Gas_432.RecTime)

Gas_432.set_index('RecTime').rolling('15T').mean()

For this dataframe, you got:

+---------------------+--------+---------+---------+---------+---------+---------+---------+----------+
| 2019-05-31 13:42:00 |  0     | 19.13   | 2225    | 4480.35 | 6503.1  | 7687.33 | -3183.6 | -8181.55 |
| 2019-05-31 13:42:00 |  8.5   | 35.385  | 1968.1  | 3967.66 | 6502.9  | 7687.09 | -3183.6 | -8181.55 |
| 2019-05-31 13:42:00 | 21.8   | 60.82   | 1774.63 | 3581.55 | 6502.57 | 7686.71 | -3183.6 | -8181.55 |
| 2019-05-31 13:42:00 | 31.375 | 79.1325 | 1624.38 | 3281.68 | 6502.27 | 7686.37 | -3183.6 | -8181.55 |
| 2019-05-31 13:42:00 | 37.88  | 91.572  | 1504.62 | 3042.69 | 6502.02 | 7686.07 | -3183.6 | -8181.55 |
+---------------------+--------+---------+---------+---------+---------+---------+---------+----------+

where the first column is datetimeIndex, the other have the same names with Gas_432.

Upvotes: 1

Related Questions