2. TimeSeries-
Meaning
A time series refers to
the values of a variable,
which are ordered
chronologically, over a
successive period of
time.
It is a sequence taken
at successive equally
spaced points in
time.
It discloses relationship
between two variables,
one being time.
2
4. Importance of Time Series Analysis
Understanding past
behavior
Predicting,
Forecasting and
Planning
Evaluation of
currentprograms
Facilitates
Comparison
4
5. Componentsof Time series
The statistical series are usually affected by a multiplicity of causes.
Such forces can be changes in population, tastes and habits of
people, changes in income etc.
The effects of these forces on the time series are called the
components of a time series
Secul
ar
Trend
Cyclic
Variation
s
Seasona
l
Variation
Irregular
Variation
s 5
6. SecularTrend
Trend refers to long
period changes.
It shows the definite
and basic tendency of
statistical data with the
passage of time.
The trend can be
Rising/Falling/Const
ant.
The concept of trend
does not include short
term oscilations.
6
13. Equation of straight line; y = a
+ bx
4.LinearTrend/Lineofbestfit.
• Here, we are trying to establish a mathematical equation and
using that equation to find the trend values.
• We are fitting a straight line to match the trend using a linear
equation derived from the available data.
Year (t) Value (y) Deviation from Mid
year; t – mid year
(x)
xy 𝒙2
𝒂=
𝒚
𝒏
𝒙
𝒚
b = 𝒙𝟐
13
17. • A seasonal index is calculated by the average of each season.
• Here, a season means a month or a quarter or any other time
period for which the fluctuations happen.
Step 1: Make a season wise arrangement of
data
Step 2: Calculate the average for each season
Step 3: Calculate the average of averages
Step 4: Calculate seasonal index using the
formula.
1.SimpleAverage Method.
Seasonal
Index=
𝑨𝒗𝒆𝒓𝒂𝒈𝒆𝒐𝒇𝒕
𝒉
𝒂
𝒕𝒔𝒆𝒂𝒔𝒐𝒏
𝑨𝒗𝒆𝒓𝒂𝒈𝒆𝒐𝒇𝒂𝒗𝒆𝒓𝒂𝒈𝒆𝒔
×
100
17
18. 2.Ratiototrendmethod
Firstly, we have to find the trend eliminated values using
method of least squares
Then method of simple average is applied on these trend
eliminated values
We will then be getting the seasonal indices
This method is recommended only when there is absence
of cyclical movements
3.RatiotoMovingaveragemethod
18
20. LorenzCurve
It is a graphic method of studying dispersion in a series
It was developed by Max O. Lorenz in 1905for
representing inequality of the wealth distribution..
It can also be used for studying the dispersion in series
such as wages, production, populationetc..
If there is no inequality in the distribution, lorenz curve
will coincide with the line of equal distribution
20