26

November 10, 2024

0.1 Exercise 26: Flights Dataset - Time Series Visualization and Analysis
[169]: import pandas as pd

[171]: df=pd.read_csv('flights.csv')

[173]: df

[173]: year month passengers

0 1949 January 112
1 1949 February 118
2 1949 March 132
3 1949 April 129
4 1949 May 121
.. … … …
139 1960 August 606
140 1960 September 508
141 1960 October 461
142 1960 November 390
143 1960 December 432

[144 rows x 3 columns]

0.2 1. Load and overview

[210]: import seaborn as sns
import pandas as pd
import matplotlib.pyplot as plt

[177]: flights = sns.load_dataset("flights")

print(flights.head())

year month passengers

0 1949 Jan 112
1 1949 Feb 118
2 1949 Mar 132
3 1949 Apr 129
4 1949 May 121

1
 0.3 2. Resample Data to Monthly
[215]: flights['month'] = flights['month'].astype(str)
flights['date'] = pd.to_datetime(flights['year'].astype(str) + '-' +␣
↪flights['month'] + '-01')

monthly_passengers = flights.groupby('date')['passengers'].mean().resample('M').
↪mean()

print(monthly_passengers)

date
1949-01-31 112.0
1949-02-28 118.0
1949-03-31 132.0
1949-04-30 129.0
1949-05-31 121.0
…
1960-08-31 606.0
1960-09-30 508.0
1960-10-31 461.0
1960-11-30 390.0
1960-12-31 432.0
Freq: M, Name: passengers, Length: 144, dtype: float64

0.4 3.Yearly Aggregation

[181]: yearly_passengers = flights.groupby('year')['passengers'].sum()
print(yearly_passengers)

year
1949 1520
1950 1676
1951 2042
1952 2364
1953 2700
1954 2867
1955 3408
1956 3939
1957 4421
1958 4572
1959 5140
1960 5714
Name: passengers, dtype: int64

2
 0.5 4.6-month rolling average
[219]: monthly_passengers_rolling = monthly_passengers.rolling(window=6).mean()
print(monthly_passengers_rolling)

date
1949-01-31 NaN
1949-02-28 NaN
1949-03-31 NaN
1949-04-30 NaN
1949-05-31 NaN
…
1960-08-31 519.166667
1960-09-30 534.000000
1960-10-31 534.000000
1960-11-30 520.333333
1960-12-31 503.166667
Freq: M, Name: passengers, Length: 144, dtype: float64

0.6 5. Line plot for Monthly Trends

[185]: import matplotlib.pyplot as plt

plt.figure(figsize=(12, 6))
plt.plot(monthly_passengers.index, monthly_passengers, label='Monthly␣
↪Passengers')

plt.plot(monthly_passengers_rolling.index, monthly_passengers_rolling,␣
↪label='6-Month Rolling Average', color='lightcoral')

plt.title('Monthly Passengers with 6-Month Rolling Average')

plt.xlabel('Date')
plt.ylabel('Number of Passengers')
plt.legend()
plt.show()

3
[186]: #Conclusion: The line plot illustrates the trend of monthly passengers over␣
↪time, showcasing fluctuations throughout the year.

#The 6-month rolling average smooths out seasonal variations and highlights an␣
↪overall upward trend in passenger numbers, indicating growing demand for air␣

↪travel.

#Peaks in passenger numbers are evident during certain months, suggesting␣

↪seasonal travel patterns.

0.7 6. Monthly Passenger Distribution

[225]: plt.figure(figsize=(12, 6))
sns.boxplot(x='month', y='passengers', data=flights)
plt.title('Monthly Passenger Distribution')
plt.xlabel('Month')
plt.ylabel('Number of Passengers')
plt.show()

4
[190]: #Conclusion: The boxplot shows the distribution of passenger counts across␣
↪different months.

#It reveals that some months have significantly higher passenger counts, while␣
↪others have lower numbers.

#The presence of outliers in certain months indicates that there are periods of␣
↪exceptionally high travel, which could be linked to holidays or vacation␣

↪seasons.

#The median line within each box provides insight into typical passenger␣
↪numbers for each month.

0.8 7. Lagged Passenger

[228]: flights['lagged_passengers'] = flights['passengers'].shift(6)
print(flights.head(12))

year month passengers date lagged_passengers

0 1949 Jan 112 1949-01-01 NaN
1 1949 Feb 118 1949-02-01 NaN
2 1949 Mar 132 1949-03-01 NaN
3 1949 Apr 129 1949-04-01 NaN
4 1949 May 121 1949-05-01 NaN
5 1949 Jun 135 1949-06-01 NaN
6 1949 Jul 148 1949-07-01 112.0
7 1949 Aug 148 1949-08-01 118.0
8 1949 Sep 136 1949-09-01 132.0
9 1949 Oct 119 1949-10-01 129.0
10 1949 Nov 104 1949-11-01 121.0

5
 11 1949 Dec 118 1949-12-01 135.0

0.9 Step 8: Trend Line for Annual Data

[231]: plt.figure(figsize=(12, 6))
plt.plot(yearly_passengers.index, yearly_passengers, marker='o',␣
↪color='hotpink')

plt.title('Yearly Passenger Trend')

plt.xlabel('Year')
plt.ylabel('Total Passengers')
plt.grid()
plt.xticks(yearly_passengers.index)
plt.show()

[233]: #The line graph shows the steady and significant growth in annual passenger␣
↪numbers between 1949 and 1960.

#It can be clearly seen that each year the number of passengers increased␣
↪compared to the previous year.

#In particular, in the late 50s, the growth rate tended to increase faster,␣
↪showing the strong development of the passenger transport industry during␣

↪this period.

#This growth can be explained by many factors, including economic development,␣

↪the expansion of transportation networks and the increase in people's income.

↪ Overall, the chart shows a very positive trend, reflecting people's growing␣

↪demand for travel and the success of the transportation industry in meeting␣

↪that demand.

6
 0.10 9.Histogram of Monthly Passengers
[244]: plt.figure(figsize=(12, 6))
plt.hist(monthly_passengers, bins=12, color='pink', alpha=0.7)
plt.title('Histogram of Monthly Passengers')
plt.xlabel('Number of Passengers')
plt.ylabel('Frequency')
plt.grid()
plt.show()

[201]: #Conclusion: The histogram provides a visual representation of the frequency␣

↪distribution of monthly passenger counts.

#It indicates that most months have passenger counts concentrated around␣
↪certain values, with fewer months experiencing very high or very low counts.

#This distribution can inform airlines about typical passenger loads, aiding in␣
↪capacity planning and resource allocation.

0.11 Step 10: Area Plot for Yearly Totals

[239]: plt.figure(figsize=(12, 6))
plt.fill_between(yearly_passengers.index, yearly_passengers, color='deeppink',␣
↪alpha=0.6)

plt.title('Area Plot for Yearly Total Passengers')

plt.xlabel('Year')
plt.ylabel('Total Passengers')
plt.grid()
plt.show()

7
[240]: #Conclusion: The area plot effectively visualizes the cumulative total of␣
↪passengers per year, emphasizing the growth trend in air travel.

#The filled area allows for a quick visual comparison between years,␣
↪highlighting significant increases or decreases in passenger counts.

#This visualization helps to convey the overall growth narrative of the airline␣
↪industry, making it easier to identify trends over time.