Time series and forecasting: Unit 2

Need for forecasting –

The time lag between awareness and occurrence of an event is the main reason for planning and
forecasting. This time lag is called as the lead time. If the lead time is zero, then there is no need for
planning and if the lead time is long then planning plays important role. In management and
administrative situations, the lead time can range from few seconds to few hours to few days to
several years. Hence, planning is of utmost important. Effective and efficient planning can be
achieved by proper forecasting.
Forecasting is an integral part of the decision-making activities of management. An organization
establishes goals and objectives, seeks to predict environmental factors, then selects actions that it
hopes will result in attainment of these goals and objectives. The need for forecasting is increasing as
management attempts to decrease its dependence on chance and becomes more scientific in
dealing with environment. Each area of an organization is related to all others; a good or bad
forecast can affect the entire organization. Hence it is required that an organization possesses
knowledge and skills covering –
1. Identification and definition of forecasting problem,
2. Application of a range of forecasting methods,
3. Procedures for selecting the appropriate methods for a situation,
4. Organizational support for applying and using formalized forecasting methods.
A forecasting system must establish linkage among forecasts made by different management areas.
There is a high degree of interdependence among the forecasts of various divisions or departments,
which cannot be ignored for successful forecasting.
One should make proper distinction between uncontrollable external events and controllable
internal events in the process of forecasting. Uncontrollable external events include those
originating with the national economy, governments, customers and competitors; while decisions
within the firm related to marketing or manufacturing are internal controllable events. The success
of a company depends upon both types of events. Forecasting directly applied to the uncontrollable
external events while decision making applies directly to the controllable events.

Forecasting techniques –
Two broad categories are as follows:
1. Quantitative Methods
 Used when situation is ‘stable’ & historical data exist
 Existing products
 Current technology
 Involve mathematical techniques
 e.g., forecasting sales of color televisions

Sufficient quantitative information should be available. Quantitative forecasting includes

a. Time Series forecasting – Predicting the continuation of historical patterns
such as the growth in sales or gross national product.
b. Explanatory Forecasting – Understanding how explanatory variables such as
prices and advertising affect sales.

Quantitative forecasting can be applied when

1. Information about the past is available,
2. The information can be quantified in the form of numerical data,
3. It can be assumed that some aspects of the past will continue into the future.
Quantitative forecasting procedures fall on a continuum between two extremes: intuitive or ad hoc
methods and formal quantitative methods based on statistical principles.

Advantages of Intuitive or ad hoc methods are

1. Simple and easy to understand.
2. Based on empirical experience that varies widely from business to business, product to product,
and forecaster to forecaster.

Limitations of Intuitive or ad hoc methods are

1. They are not as accurate as formal quantitative methods
2. They usually give little or no information about the accuracy of the forecast.

Due to these limitations of intuitive methods, its use has declined and formal methods have gained
popularity. Formal statistical methods can involve extrapolation, but it is done in a standard way
using systematic approach that attempts to minimize the forecasting error.

Explanatory models assume that the variable to be forecasted exhibits an explanatory relationship
with one or more independent variables. Explanatory models can be applied to systems explaining
national economy, company’s market or a household. The purpose of explanatory model is to
discover the form of the relationship and use it to forecast future values of the forecast variable. Any
change in the inputs will affect the output of the system in a predictable way assuming that the
explanatory relationship will not change.
Unlike explanatory forecasting, time series forecasting treats the system as a black box and makes
no attempt to discover the factors affecting its behaviour.
Hence, in time series forecasting, the future is based on past values of a variable and/ or errors. The
objective of such time series forecasting methods is to discover the pattern in the historical data
series and extrapolate that pattern into the future.

2. Qualitative Methods
 Used when situation is vague & little data exist
o New products
o New technology
 Involve intuition, experience
 e.g., forecasting sales on Internet, forecasting the speed of
telecommunications around the year 2020
It is difficult to measure the usefulness of qualitative forecasts. They are used mainly to provide
hints, to aid the planner and to supplement quantitative forecasts rather that to provide a specific
numerical forecast.

Basic steps in forecasting are:

1. Problem definition
2. Gathering information
3. Preliminary (exploratory) analysis
4. Choosing and fitting models
5. Using and evaluating a forecasting model.
 Time series data, analysis and importance

1. Definition of Time series data:

Time series data: A set of data depending on the time

Or

A series of values varying over time

A time series data consists of a set of observations measured at specified, usually equal, time
interval.

Time: Year, months, days, weeks, hours, minutes, seconds

2. Mathematical presentation of Time series:

A time series is a set of observation taken at specified times, usually at ‘equal intervals’.
Mathematically a time series is defined by the values Y1, Y2…of a variable Y at times t1, t2…. Thus,

Y= F(t)
 3. Time series Analysis:

Time series analysis attempts to identify those factors that exert influence on the values in the
series. Time series analysis is a basic tool for forecasting. Industry and government must forecast
future activity to make decisions and plans to meet projected changes. An analysis of the trend of
the observations is needed to acquire an understanding of the progress of events leading to
prevailing conditions.

4. Time series examples:

• Sales data
• Gross national product
• Share prices
• $A Exchange rate
• Unemployment rates
• Population
• Foreign debt
• Exchange rate, interest rate, inflation rate, national GDP
• Electric power consumption
• Number of accident fatalities

5. Importance of time series:

• A very popular tool for Business Forecasting.

• Basis for understanding past behavior.
• Can forecast future activities/planning for future operations
• Evaluate current accomplishments/evaluation of performance.
• Facilitates comparison
• Estimation of trade cycle

Components of time series:

1. What are components of time series?
The characteristics movement or fluctuation in time series is referred to as component of
time series. These fluctuations or variations can occur for various reasons.

2. Causes of variations in time series:

• Social customs, festivals etc.
• Seasons
• The four phase of business : prosperity, decline, depression, recovery
• Natural calamities: earthquake, epidemic, flood, drought etc.
 • Political movements/changes, war etc.

3. Time series components:

Time series data can be broken into these four components:

• Secular trend: long term pattern of a time series (increasing or decreasing)

• Seasonal variation: are yearly periodic pattern in a time series repeated on a regular basis
• Cyclical variation: refer to a business cycle that lasts for 2 to 10 years
• Irregular variation: are abnormal time series movements occurring due to unforeseen
events.

4. Description of different components:

a. Trend or Secular trend (long term variation):

The secular trend is the main component of a time series which results from long term
effects of socio-economic and political factors. This trend may show the growth or decline in
a time series over a long period. This is the type of tendency which continues to persist for a
very long period. It describes the history of time series. Prices and export and import data,
for example, reflect obviously increasing tendencies over time
Other examples include increase in prices, increase in pollution, increase in the need of
wheat, increase in literacy rate, decrease in deaths due to advances in science.
The term ‘long term’ is used subjectively. Generally, in economic terms, long term may
mean >10 years. But for example, for the time series measuring the bacterial growth per
minute, duration of 2 hours could be sufficiently large where as for the quarterly sales of a
company, two years could be short duration.
Detection of trend: Taking averages over a certain period is a simple way of detecting trend
in seasonal data. Change in averages with time is evidence of a trend in the given series,
though there are more formal tests for detecting trend in time series.
Purpose of measuring Trend:
 Knowledge of past behaviour
 Estimation
 Study of other components
Example:
One observes an increasing trend in the following time series plot of yearly revenue.
 b. Seasonal effect (seasonal variation or fluctuation):
These are short term movements occurring in data due to seasonal factors. The short term
is generally considered as a period in which changes occur in a time series with variations in
weather or festivities. Seasonal Fluctuations describes any regular variation (fluctuation)
with a period of less than one year.
For example, it is commonly observed that the consumption of ice-cream during summer is
generally high and hence an ice-cream dealer’s sales would be higher in some months of the
year while relatively lower during winter months. Employment, output, exports, etc., are
subject to change due to variations in weather. Similarly, the sale of garments, umbrellas,
greeting cards and fire-works are subject to large variations during festivals like Valentine’s
Day, Eid, Diwali, Christmas, New Year's, etc.
These types of variations in a time series are isolated only when the series is provided
biannually, quarterly or monthly.
Seasonal variations are usually due to the Climate & weather condition and Customs
traditions & habits like festive occasion.
Examples include:
 Air conditioner sales in Summer
 Heater sales in Winter
 Flu cases in Winter
 Airline tickets for flights during school vacations
 Fire crackers sales during Diwali

The changes which repeat themselves within a fixed period, are also called seasonal
variations, for example, traffic on roads in morning and evening hours, Sales at festivals like
EID etc., increase in the number of passengers at weekend etc.
Systematic seasonal fluctuations are clearly visible in the following plot of the monthly CO2
concentration data.

Characteristic of seasonal variation:

• Regularity
• Fixed proportion
• Increase or Decrease
• Easy forecast

Purpose of seasonal variation:

• Analysis of past behavior of the series
• Forecasting the short time fluctuations
• Elimination of the seasonal variations for measuring cyclic variations

c. Cyclical changes (cyclical variations or cyclical fluctuations:

These are long term oscillations occurring in a time series. These oscillations are mostly observed in
economics data and the periods of such oscillations are generally extended from five to twelve years
or more. These oscillations are associated with the well-known business cycles. These cyclic
movements can be studied provided a long series of measurements, free from irregular fluctuations,
is available.
Cyclic fluctuations/variations are long term movements that represent consistently recurring rises
and declines in activity. The name is quite misleading because these cycles can be far from regular
and it is usually impossible to predict just how long periods of expansion or contraction will be.
There is no guarantee of a regularly returning pattern.

Cause: Time series exhibits Cyclical Variations at a fixed period due to some other physical cause,
such as daily variation in temperature. Cyclical variation is a non-seasonal component which varies in
recognizable cycle. sometime series exhibits oscillation which do not have a fixed period but are
predictable to some extent. For example, economic data affected by business cycles with a period
varying between about 5 and 7 years. In weekly or monthly data, the cyclical component may
describe any regular variation (fluctuations) in time series data.

Example include:

• Changes in interest rates

• Economic depressions or recessions

• Changes in consumer spending

The cyclical variation is periodic in nature and repeat themselves like business cycle, which has four
phases (i) Peak/ Prosperity (ii) Recession (iii) Trough/Depression (iv) Recovery.

• Purpose of studying cyclical variation: Useful in formulating policies in business

d. Irregular variation or Random component (Irregular fluctuation)

These are sudden changes occurring in a time series which are unlikely to be repeated. They
are components of a time series which cannot be explained by trends, seasonal or cyclic
movements. These variations are sometimes called residual or random components. It
follows no pattern and is by nature unpredictable.These variations, though accidental in
nature, can cause a continual change in the trends, seasonal and cyclical oscillations during
 the forthcoming period. Floods, fires, earthquakes, revolutions, epidemics, strikes etc., are
the root causes of such irregularities. Irregular variation cannot be explained
mathematically.

When trend and cyclical variations are removed from a set of time series data, the residual left,
which may or may not be random. Various techniques for analyzing series of this type examine to
see “if irregular variation may be explained in terms of probability models such as moving
average or autoregressive models, i.e. we can see if any cyclical variation is still left in the residuals.
These variations occur due to sudden causes are called residual variation (irregular
variation or accidental or erratic fluctuations) and are unpredictable, for example rise in prices of
steel due to strike in the factory, accident due to failure of break, flood, earth quake, war etc.

Characteristic of random component:

• Short period of time

• No Statistical technique

• Unpredictable

Recent example:
No. of air passengers over years has a sudden decrease in graph due to Corona outbreak.

Decomposition of Time series

1. Analysis of time series data:

• Obtain an understanding of the underlying forces and structure that produced the observed
data.
• Time Series models attempt to identify significant patterns in the components of a time
series.
• Assuming that these patterns continue into future, time series models extrapolate these
patterns to forecast future time series values.
• Fit a model and proceed to forecasting.

2. Meaning of decomposition:
Yt = f(Tt ,St ,Ct, Rt)

where

• Yt = data at period t
• Tt = trend-cycle component at period t
• St = seasonal component at period t
• Ct = cyclical component at period t
• Rt = remainder component at period t

Additive model:
Yt = St + Tt + Ct + Rt .
In most of the cases, Yt = St + Tt + Rt
Multiplicative model:
Yt = St × Tt × Ct × Rt
In most of the cases, Yt = St × Tt × Rt

3. Concept of decomposition:
Macroeconomic time series are subject to two sorts of forces: those that influence the long-
run behavior of the series and those that influence the short-run behavior of the series.
Long-run behavior: Trend
Short-run behavior: Variations in data
De-trended and de-seasonalised data has to be obtained. There are various methods of
decomposition, easiest among them is classical decomposition.

4. Steps of decomposition:
The classical decomposition method originated in the 1920s. It is a relatively simple
procedure, and forms the starting point for most other methods of time series
decomposition. There are two forms of classical decomposition: an additive decomposition
and a multiplicative decomposition. In classical decomposition, we assume that the seasonal
component is constant from year to year.

 Calculate trend values

 Calculate de-trended values
 Estimate seasonal component
 Remainder component is calculated by subtracting (additive model) or dividing
(multiplicative model) the trend and seasonal variation from the original data.

5. Forecasting and decomposition:

While decomposition is primarily useful for studying time series data, and exploring
historical changes over time, it can also be used in forecasting.

Trend can be forecasted by method of Least squares. Also, To forecast a decomposed time
series, we forecast the seasonal component and the seasonally adjusted component
separately.