Shifting Time Series Data

In time series analysis, shifting is a process of re-arranging time series data with some lag $n$ typically inorder to perform comparison (Year over Year, Month over Month) or implement a differencing operation.

Generating time series data

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

ts_index = pd.date_range( '2013-01-01', '2021-01-01', freq='Y' )
data = np.random.randint(56, 109, size=8)

ts_data = pd.DataFrame(data=data, index=ts_index, columns=['yearly_sales'])
ts_data

Implementing Shift Operation

ts_data['last_year_sales'] = ts_data['yearly_sales'].shift(1)
ts_data