Autocorrelation plots

Autocorrelation plots graph autocorrelations of time series data for different time lags. In layman's terms, autocorrelation is the correlation of the values at time n and the values at time n+l, where l is the time lag. Generally, these plots are used for checking whether the time series has randomness in its progression. Autocorrelations are near zero for all time-lag separations in the case of a random time series, and have a non-zero value of significance at some or all time-lag separations for a non-random time series. We explain autocorrelation further in Chapter 7, Signal Processing and Time Series.

The autocorrelation_plot() Pandas function in pandas.tools.plotting can draw an autocorrelation plot. The following is the code from the ch-06.ipynb file in this book's code bundle:

import matplotlib.pyplot as plt 
import numpy as np 
import pandas as pd 
from pandas.tools.plotting import autocorrelation_plot 
 
df = pd.read_csv('transcount...