Write a program in Python to print the first and last three days from a given time series data

When working with time series data in Pandas, you often need to extract specific time periods. The first() and last() methods allow you to retrieve data from the beginning and end of a time series based on a time offset.

Creating Time Series Data

First, let's create a time series with city names indexed by dates ?

import pandas as pd

# Create a series with city names
data = pd.Series(['Chennai', 'Delhi', 'Mumbai', 'Pune', 'Kolkata'])

# Create a date range with 2-day frequency
time_series = pd.date_range('2020-01-01', periods=5, freq='2D')

# Set the date range as index
data.index = time_series

print("Time series data:")
print(data)

Time series data:
2020-01-01    Chennai
2020-01-03      Delhi
2020-01-05     Mumbai
2020-01-07       Pune
2020-01-09    Kolkata
Freq: 2D, dtype: object

Extracting First and Last Three Days

Use the first() and last() methods with a time offset to get data from specific periods ?

import pandas as pd

data = pd.Series(['Chennai', 'Delhi', 'Mumbai', 'Pune', 'Kolkata'])
time_series = pd.date_range('2020-01-01', periods=5, freq='2D')
data.index = time_series

# Get first three days of data
first_three_days = data.first('3D')
print("First three days:")
print(first_three_days)

print("\n" + "="*30 + "\n")

# Get last three days of data
last_three_days = data.last('3D')
print("Last three days:")
print(last_three_days)

First three days:
2020-01-01    Chennai
2020-01-03      Delhi
Freq: 2D, dtype: object

==============================

Last three days:
2020-01-07    Pune
2020-01-09    Kolkata
Freq: 2D, dtype: object

How It Works

The first() and last() methods work with time offsets:

• first('3D'): Returns data from the start date up to 3 days later
• last('3D'): Returns data from 3 days before the end date to the end
• Time offset '3D': Represents a 3-day period

Alternative Approaches

You can also use head() and tail() to get a specific number of records ?

import pandas as pd

data = pd.Series(['Chennai', 'Delhi', 'Mumbai', 'Pune', 'Kolkata'])
time_series = pd.date_range('2020-01-01', periods=5, freq='2D')
data.index = time_series

# Get first 2 records
print("First 2 records using head():")
print(data.head(2))

print("\n" + "="*30 + "\n")

# Get last 2 records
print("Last 2 records using tail():")
print(data.tail(2))

First 2 records using head():
2020-01-01    Chennai
2020-01-03      Delhi
Freq: 2D, dtype: object

==============================

Last 2 records using tail():
2020-01-07    Pune
2020-01-09    Kolkata
Freq: 2D, dtype: object

Conclusion

Use first() and last() methods with time offsets like '3D' to extract time-based subsets from time series data. For record-based extraction, use head() and tail() methods instead.