Scribd
Upload
93 views2 pages

Time Series Analysis

This document discusses time series analysis. It explains that a time series is a set of observations made over time involving the same variable. The document outlines the key components of a time series including trend, seasonality, cyclical patterns, and irregular components. It also discusses the importance of time series analysis for understanding past behavior, planning for the future, and evaluating performance over time.

Uploaded by

midori
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
93 views2 pages

Time Series Analysis

This document discusses time series analysis. It explains that a time series is a set of observations made over time involving the same variable. The document outlines the key components of a time series including trend, seasonality, cyclical patterns, and irregular components. It also discusses the importance of time series analysis for understanding past behavior, planning for the future, and evaluating performance over time.

Uploaded by

midori
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

12/2/2017

Learning Objectives
 Explain the components of a time series
 Explain a time series model and the underlying
structure of a time series
 Understand the additive and multiplicative models
for time series
Time Series Analysis  Calculate, interpret and use the trend
 Calculate, interpret and use the seasonal
component
 Calculate, interpret and use the residual
component
 Carry out elementary forecasting from time
series analysis

Introduction
 A univariate time series is a ordered set of
measurements or observations of the same
variable made through time (daily, weekly, monthly,
quarterly, yearly..etc), for which the time aspect is
potentially important to understanding; an
important characteristic of the measurements,
that are made at regular time intervals, is their
order.

 One such example is the number tourists that


visited Mauritius from 2009 to 2011. The data is
collected during every quarter of the year (i.e.
data is collected four times a year) and it is shown
in the table below.

Time Series data


 If we observe the above series, we find that generally the tourist
arrivals have increased but for some quarters a decline is also noticed.
There may be several causes responsible for increase or decrease
from one period to another such as preferred climate, changes in
tastes and habits of tourists, and availability of alternative
destinations…etc.

 It may be difficult to study the effect of various factors that have led
either to an increase or decrease in sales. However, we can analyze the
effect of the various forces under four main points

1
12/2/2017

Components of a Time Series Components of a Time Series


 Trend Component: ignoring short-term  Seasonality Component: Seasonality refers
fluctuations, the measurements tend to increase to regularly repeating pattern of highs and lows
(or decrease) over long period of time? related to calendar time such as seasons,
quarters, months, days of the week, and so on.

 E.g. Number tourists visiting Mauritius tend to  For instance number of tourists visiting
increase; there were 871356 tourists in 2009 Mauritius in the fourth quarter is definitely
and 964642 in 2011. This is called an upward higher than in other quarters. The seasonality
trend. component is important if we want to forecast
each quarter as precisely as possible.

Components of a Time Series Components of a Time Series


 Cyclical Component: Is there a long-run cycle or  Irregular component: Are there any abrupt
period unrelated to seasonality factors? changes to either the level of the series or the
variance?
 Changes that have taken place as a result of booms  It is also important to identify outliers (outliers
and depressions. are data located far away from other data).

 Changes that have taken place as a result of


 Thus, seasonal patterns are not the only regular and such forces that could not be predicted like
oscillatory behaviour; cyclical patterns are also floods, earthquakes, famines…etc. such changes
oscillatory but are observed when we have data for are classified as ‘Irregular or Erratic
a very long period (e.g. several decades). Variations (Residual).

Role of Time Series Analysis Time Series


 A time series is a sequence of data points, measured
 Time series analysis is of great significance in typically at successive times spaced at uniform time
business decision making for the following intervals.
reasons:
 A time series data set shows how a variable changes over
 It helps in understanding of past behaviors. time.
 It helps in planning future operations
◦ Examples:
 It helps in evaluating current accomplishments  Price series of petrol over the last 10 years
 It facilitates comparison. Different time series  Quarterly sales figures for a company over a period of time
 Monthly GDP of a country
are often compared and important conclusions
drawn.  In a time series plot, the X-axis is always the time axis
and the Y-axis will have the data values.

You might also like