Time Series Analysis & Modeling

with the Airport Passengers Dataset

STAT 437
Xinyuan Gao & Chenjia Hui
Introduction

 Data Overview
 1. Perform Exploratory Data Analysis (EDA)
2. Time Series Decomposition
3. Test the Stationarity
4. Fit a Model used ARIMA Algorithm
5. Calculate Forecasts
 Takeaway
 Data Overview

https://data.world/data-society/air-traffic-passenger-data
 1. Exploratory Data Analysis

 Passenger numbers had fallen from 2008 to 2010.

 1. Exploratory Data Analysis(cont.)

 The means of passengers travelling in months 6 to 9 are higher than

means of passengers travelling in other months.
 2. Time Series Decomposition

The multiplicative model is: Y[t]=T[t]∗S[t]∗e[t]

where
Y(t) is the number of passengers at time t,
T(t) is the trend component at time t,
S(t) is the seasonal component at time t,
e(t) is the random error component at time t.
2. Time Series Decomposition(cont.)
 Trend: The long-term general change in the level of the
data with a duration of longer than a year.
 Seasonal variations: Regular wavelike fluctuations of
constant length, repeating themselves within a period of
no longer than a year.
 Cyclical variations: Wavelike movements, quasi
regular fluctuations around the long-term trend, lasting
longer than a year.
 2. Time Series Decomposition(cont.)

 Decomposing the data into its trend, seasonal, and random error components will give
some ideas how these components relate to the observed dataset.
3. Test Stationarity of the Time Series

 a) Augmented Dickey-Fuller (ADF) Test

H0: The time series is non-stationary
Ha: The time series is stationary
 3. Test Stationarity of the Time Series (cont.)

 b) Autocorrelation
- measures the
dependency between
variables in a series

Stationarity - the
residuals are centered
around 0.
4. Fit a Time Series Model
 ARIMA Model (Autoregressive Integrated Moving Average)
ARIMA (p,d,q)(P, D, Q)[m]
p:  the number of autoregressive
d: degree of differencing
q: the number of moving average
terms
m: refers to the number of periods
in each season
(P, D, Q): represents the (p,d,q) for
the seasonal part of the time series
4. Fit a Time Series Model (cont.)
5. Time Series Forecast
Takeaway

 Adjust price
 Improve passengers’ experience
(seats, food, entertainment, etc. )
Thank You!