You are on page 1of 17

BY

DR. RAKSONI GUPTA


ASSISTANT PROFESSOR
DEPARTMENT OF STATISTICS
drraksonigupta@gmail.com
Time Series
Definition
❖ Arrangement of statistical data in chronological order, i.e., in
accordance with occurrence of time is called a time series. It have an
unique importance in the field of Economics and Business statistics.
❖ Mathematically ,time series is defined by functional relationship
U t = f (t )
where U is the value of phenomena under consideration at different
t
time t
For example:
1. Population of a country in different years t
2. The temperature of a place at different days of a week and so on
Thus if the values of the phenomena at time t1 , t2 ,...., tn are U1 ,U 2 ,...,U n
then the series thus generated
t : t1 , t 2 , ...t n
constitutes a time series U t : U1 , U 2 , ..., U n
Components of time series

The various forces at work affecting the value of phenomena in a


time series can be broadly classified into following four
categories, commonly called as components of a time series.
COMPONENTS OF TIME SERIES
Basic
components of
time series

Long term Short term


fluctuations fluctuations

Random
Secular
Cyclic Seasonal component or
fluctuations or
fluctuations fluctuations irregular
trend
fluctuations
1. Secular fluctuation or Trend
❖ Secular trend or simply Trend, we mean the general tendency of
data decrease during a long period of time, the trend may be
upward or downward example the data pertaining to population
agricultural production is an example of upward trend. On the
other hand the infant mortality rate which is constantly
decreasing is an example of downward Trend.
❖ Linear or nonlinear curvilinear Trend : If time series values
plotted on a graph clusters more or less around a straight line
then the trend exhibited by time series is termed as linear
otherwise it is non-linear.
2. Seasonal variation
❖ The variations in a time series are due to rhythmic forces which operate in a
regular and periodic manner over a span of less than i.e. during a period of 12
months and have the same or almost same pattern year after year. Thus,
seasonal variation in a time series will be there if the data are recorded
quarterly, monthly, weekly, daily and so on. The seasonal variation be
attributed to the following two causes:
❖ Those resulting from natural causes: As the name suggests the various Seasons
or weather conditions and climatic changes plays an important role in seasonal
movement. For example: the sale of umbrella pick up very fast in rainy season,
the demand for electric fan goes up in summer, the sale of ice cream increases
very much in summer, the sale of woollen clothes goes up in winter season, all
be affected by natural forces i.e. weather or season.
❖ Those resulting from manmade conventions: These variations in time series
within period of 12 months are due to habits passion custom or conventions of
the people in the society for example the sale of jewellery goes up in marriages,
the sales and profits in departmental stores goes up considerable high during
festivals like Diwali and Dussehra etc.

3. Cyclic variation
❖ The oscillatory movement in time series with period of
oscillation more than one year is termed as cyclic fluctuation. A
complete period is called a cycle. The cyclic movement in a time
series are generally attributed to the so-called business cycle
which may also refer to as the four phase cycle composed of
prosperity (period of Boom), recession, depression and recovery
and it normally last for 7 to 11 years.
4. Random movement
❖ These fluctuations are purely random, erratic, unforeseen,
unpredictable and are due to numerous non-recurring and
irregular circumstances which are beyond the control of human
hand but are sometime are a part of our system such as epidemic,
war, revolution etc.
Analysis of Time Series
The main problem in a time series is to identify the forces or
component at work, the net effect of these components is
exhibited by the movement of time series, we try to isolate,
study, analyse and measure them independently i.e. by holding
other things constant.
Mathematical models for time
series
❖ Decomposition by Additive model
According to this model time series can be represented by
Ut = Tt + St + Ct + Rt
where Ut is the time series value at time t.
Tt represents the Trend values, St represent the seasonal, Ct
represent the cyclic and Rt is the random variation
❖Decomposition by Multiplicative model
According to multiplicative model
Ut = Tt * St * Ct * Rt
where St ,Ct , Rt instead of assuming positive and negative values,
these are indices fluctuating above and below unity. Geometric
mean of St in year, Ct in cycle and Rt in long term period are Unity.

You might also like