Python Rolling Mean

Intro

When working with time series, we often want to view the average over a certain number of days. For example, we can view a 7-day rolling average to give us an idea of change from week to week. In this article, we will learn how to conduct a moving average in python.

Data

Let's load a data set of monthly milk production. We will load it from the url below. The data consists of monthly intervals and kilograms of milk produced.

import pandas as pd

df = pd.read_csv('https://raw.githubusercontent.com/pandas-dev/pandas/master/doc/data/air_quality_no2_long.csv')

df.head()

date_col = 'date.utc'
df = df[[date_col, 'value']]


df[date_col] = pd.to_datetime(df[date_col])
df = df.set_index(date_col)
df = df.sort_index()

C:\Users\krh12\AppData\Local\Temp/ipykernel_13784/1677584757.py:5: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  df[date_col] = pd.to_datetime(df[date_col])

df.head()

df.plot()

<AxesSubplot:xlabel='date.utc'>

Conducting a moving average

To conduct a moving average, we can use the rolling function from the pandas package that is a method of the DataFrame. This function takes three variables: the time series, the number of days to apply, and the function to apply. In the example below, we run a 2-day mean (or 2 day avg).

twoday_mean = df.rolling('2D').mean()

We can also plot the data over our orignal time series to see how the avg smoothed out the data.

import matplotlib.pyplot as plt

fig, ax = plt.subplots(1,1,figsize=(10,5))

ax.plot(df.value)
ax.plot(twoday_mean)

plt.show()

Let's do another example with a 7-day avg which is a common task in disease outbreaks and stocks.

sevenday_mean = df.rolling('7D').mean()
sevenday_mean

import matplotlib.pyplot as plt

fig, ax = plt.subplots(1,1,figsize=(10,5))

ax.plot(df.value)
ax.plot(sevenday_mean)

plt.show()

Other Rolling Functions

You may have noticed from the above that we can do more than a rolling average with the rolling function. We can actually apply any math function. Let's run a couple of more examples, sum and median.

df.rolling('7D').median()

df.rolling('7D').sum()