Академический Документы
Профессиональный Документы
Культура Документы
LAW SCHOOL
B.COM.LL.B. (HONS.),
FIFTH SEMESTER
2016
PROJECT BY:
K.VISHAL
BCO140066
Declaration
I, K.VISHAL do hereby declare that the project entitled Analysis of Time Series
submitted to Tamil Nadu National law school in partial fulfilment of requirement of award of
degree in undergraduate in law is a record of original work done by me under the supervision
and guidance of Dr. T.S. Agilla, Asst. Professor of Law Tamil Nadu National law school and
has not formed basis for award of any degree or diploma or fellowship or any other title to
any other candidate of any university.
K.VISHAL
B.Com, LL.B (Hons)
ACKNOWLEDGEMENTS
At the outset, I take this opportunity to thank my Professor Dr. T.S. Agilla who has been of
immense help during moments of anxiety and torpidity while the project was taking its
crucial shape.
Secondly, I convey my deepest regards to the Vice Chancellor and the administrative staff of
TNNLS who held the project in high esteem by providing reliable information in the form of
library infrastructure and database connections in times of need.
Thirdly, the contribution made by my parents and friends by foregoing their precious time is
unforgettable and highly solicited. Their valuable advice and timely supervision paved the
way for the successful completion of this project.
Finally, I thank the Almighty who gave me the courage and stamina to confront all hurdles
during the making of this project.
Table of Contents
1.
INTRODUCTION................................................................................................ 5
1.1 AIMS & OBJECTIVES........................................................................................ 5
1.2. SCOPE & SIGNIFICANCE OF THE STUDY...........................................................5
1.3. RESEARCH METHODOLOGY............................................................................5
2. Introduction to Statistics........................................................................................... 6
2.1 Meaning and Definition of Statistics.........................................................................6
3. Time Series Analysis- An Introduction...........................................................................7
3.1 Trends, Seasonality, Cycles and Residuals..................................................................9
3.2 Utility of Time Series Analysis...............................................................................9
4. Components of Time Series...................................................................................... 10
4.1 Multiplicative Model.......................................................................................... 11
4.2 Additive Model................................................................................................. 11
5. Measurements of Secular Trend.................................................................................12
5.1 Freehand Method................................................................................................. 12
5.2 Method of Averages.............................................................................................. 13
5.2.1 Moving Averages............................................................................................ 13
5.2.1.1. Odd and Even Number of Years....................................................................15
5.2.2. Weighted Moving Averages..............................................................................17
5.2.3. Semi-Average Method..................................................................................... 19
6. CONCLUSION..................................................................................................... 23
BIBLIOGRAPHY..................................................................................................... 24
1. INTRODUCTION
1.1 AIMS & OBJECTIVES
The researcher aims to study the concept of time series analysis by studying about the
utility, different variations and the components of time series.
The researcher would also like to study the two different models of representation of
time series- the Additive Model and the Multiplicative Model.
The researcher through illustrations will study the application of the moving averages,
weighted averages and the semi-average method in calculating time series values.
1.2. SCOPE & SIGNIFICANCE OF THE STUDY
This research project will help in a study of how statistical data is affected by time series
components like trends, seasonal or cyclical variations. The researcher feels the true
importance of such study is because the research project focuses on the additive model of
representation of time series problems and the illustrations provided in better understanding
the moving averages and weighted averages method of solving and organizing statistical data.
This project is limited only to the additive model of representation of secular trend
and time series analysis. It does not seek to explain the multiplicative model. This project is
made on both primary and secondary data and some of the procuring of primary data was also
restricted due to constraints of time and resources.
1.3. RESEARCH METHODOLOGY
This is a doctrinal research project and the relevant material for this project has been
collected from primary and secondary sources. The doctrinal research is a type of research
based on principles and propositions made earlier. It is based largely on secondary sources
like books, and various websites. For the purpose of the said research project the researcher
has collected the relevant material form books on statistics and specifically articles and books
on measures of dispersion.
2. Introduction to Statistics
One of the most important tasks before the economists and businessmen these days is to make
estimates for the future. For example, a businessman is interested in finding out his likely sales in the
year 2016 or as a long term planning in 2020 or the year 2030 so that he could adjust his production
accordingly and avoid the possibility of either unsold stock or inadequate production to meet the
demand. Similarly an economist is interested in estimating the likely population in the coming year so
that proper planning can be carried out with regard to food supply, jobs for people, etc. However the
first step in making estimates for the future consists of gathering information from the past, in this
connection one usually deals with statistical data which are collected, observed or recorded at
successive intervals of time. Such data are generally referred to as time-series. Thus when we
observe numerical data at different points of time the set of observations is known as time series. For
example, if we observe production, sales, population, import, export, etc. at different points, say over
the last 5 or 10 years, the set of observations formed shall constitute time series. Hence in the analysis
of the time series, time is the most important factor because the variable is related to time which may
be either year, month, week, day, hour or even minutes or seconds.
collective, natural or social phenomena from the results obtained by the analysis or
enumeration or collection of estimates. Seligman explored that statistics is a science that deals
with the methods of collecting, classifying, presenting, comparing and interpreting numerical
data collected to throw some light on any sphere of enquiry. Spiegal defines statistics
highlighting its role in decision-making particularly under uncertainty, as follows: statistics is
concerned with scientific method for collecting, organising, summa rising, presenting and
analyzing data as well as drawing valid conclusions and making reasonable decisions on the
basis of such analysis.
Accepted definitions:-
Patterson-A time series consists of statistical data which are collected, recorded, observed
over successive increments.
It is clear from the above definitions that time series consists of data arranged chronographically. Thus
if we record the data relating to population, per capita income, prices, production, etc., for the last
5,10,15,20 years or some other time period, the series so emerging would be called time series.
It should be noted that the term time series is usually used with reference to economic data and the
economists are largely responsible for the development of the techniques of time series analysis.
However the term time series can apply to all other phenomenon that are related to time such as the
number of accidents that occur in a day, the variation in the temperature of a patient during a certain
period, number of marriage taking place during a certain period, etc.
Statistics should be placed in relation to each other. If one collects data unrelated to each
other, then such data will be confusing and will not lead to any logical conclusions. Data
should be comparable over time and over space3.
2 Boddington, Statistics and their Application to Commerce, H.F. Lynch & Co. Ltd., Cornell
University, (1921)
3 Bloomfield .P, Fourier analysis of time series: An introduction. New York: Wiley.
3.Utility of Time Series:The analysis of time series is of great significance not only to the economist and businessman but also
to the scientist, astronomist, geologist, sociologist, biologist, researcher, worker, etc. for the following
reasons:
It helps in understanding past behavior. By observing data over a period of time one can
easily understand what changes have taken place in the past. Such analysis will be extremely
future variations would become possible. Thus time series helps in planning future operations.
It helps in evaluating current accomplishments. The actual performance can be compared with
not foretellers.
When this time series is implemented in a business one can undoubtedly improve
substantially in conducting his business better as he will be better prepared with the scientific
estimation which is bound to happen more than guesswork.
The idea is to create separate models for these four elements and then combine them, either
additively X = T + I + C + E or multiplicatively X = T I C E.
If we study irregular curve year by year, we see that in each year the curve starts with a low
figure and reaches a peak about the middle of the year and then decreases again. This type of
fluctuation, which completes the whole sequence of changes within the span of a year and has
about the same variation, is called seasonal variation. Furthermore, looking at the broken
curve superimposed on the original irregular curve, we find pronounced fluctuations moving
up and down every few years throughout the length of the chart. These are known as business
cycles or cyclical fluctuations. They are so called because they comprise a series of repeated
sequence just as a wheel goes round and round. Finally, the little saw tooth irregularities on
the original curve represent what are referred to as irregular movements4.
Additionally, time series analysis techniques may be divided into parametric and non
parametric methods. The parametric approaches assume that the underlying stationary
stochastic processes have a certain structure which can be described using a small number of
parameters (for example, using an autoregressive or moving average model). In these
approaches, the task is to estimate the parameters of the model that describes the stochastic
process. By contrast, non-parametric approaches explicitly estimate the covariance or
the spectrum of the process without assuming that the process has any particular structure.
3.2 Utility of Time Series Analysis
Analysis of time series is of great significance not only to the economist and businessman but
also to the scientist, astronomist, geologist, sociologist, biologist research worker, etc.
It helps in understanding past behavior. By observing data over a period of time one
can easily understand what changes have taken place in the past. Such analysis will
be extremely helpful in predicting the future behavior.
It helps in planning future operations. Plans for the future cannot be made without
forecasting events and the relationship they will have. Statistical techniques have
been evolved which enable time series to be analyzed in such a way that the
influences which have determined the form of that series may be ascertained. If the
4 Supra note 3 (S.P. Gupta)
regularity of occurrence of any feature over sufficiently long period could be clearly
established then, within limits predication of probable future variations would
become possible. Thus time series analysis helps us to cope with uncertainty about
the future.
It helps in evaluating current accomplishments. The actual performance can be
compared with the expected performance and the cause of variation analysed. If
expected sale for 2013-14 was 10000 refrigerators and the actual sale was only 9000
one can investigate the cause of the shortfall in achievement. Time Series Analysis
will enable us to apply the scientific procedure of holding other things constant as
we examine one variable at a time. For example, if we know how much the effect of
seasonality on business is we may devise ways and means of ironing out the seasonal
influence or decreasing it by producing commodities with complementary seasons.
It facilitates comparison. Different time series are often compared and important
conclusions are drawn therefrom. However, one should not be led to believe that by
time series analysis one can foretell with 100 percent accuracy that course of future
events. After all statisticians are not foretellers. This could be possible only if the
influence of the various forces which affect these series such as climate, customs and
traditions, growth and decline factors and the complex forces which produce business
cycles would have been regular in their operation. However, the facts of life reveal
that this type of regularity does not exist. But this then does not mean that time series
analysis is of no value. When such analysis is coupled with a careful examination of
current business indicators one can undoubtedly improve substantially upon
guestimates (estimates based upon guesswork) in forecasting future business
conditions5.
10
The purpose of decomposition models is to break a time series into its components: Trend
(T), Cyclical (C), Seasonality (S), and Irregularity (I). Decomposition of time series provides
a basis for forecasting. There are many models by which a time series can be analysed; two
models commonly used for decomposition of a time series are discussed below.
11
The time series methods are concerned with taking some observed historical pattern for some
variable and projecting this pattern into the future using a mathematical formula. These
methods do not attempt to suggest why the variable under study will take some future value.
This limitation of the time series approach is taken care by the application of a causal
method. The causal method tries to identify factors which influence the variable is some way
or cause it to vary in some predictable manner. The two causal methods, regression analysis
and correlation analysis, have already been discussed previously.
A few time series methods such as freehand curves and moving averages simply describe
the given data values, while other methods such as semi-average and least squares help to
identify a trend equation to describe the given data values.
7 Supra note 8
12
(iii) It is very time-consuming to construct a freehand trend if a careful and conscientious job
is to be done8.
13
In this method, the term moving is used because it is obtained by summing and averaging
the values from a given number of periods, each time deleting the oldest value and adding a
new value.
The limitation of this method is that it is highly subjective and dependent on the length of
period chosen for constructing the averages. Moving averages have the following three
limitations:
(i) As the size of n (the number of periods averaged) increases, it smoothens the variations
better, but it also makes the method less sensitive to real changes in the data.
(ii) Moving averages cannot pick-up trends very well. Since these are averages, it will always
stay within past levels and will not predict a change to either a higher or lower level.
(iii) Moving average requires extensive records of past data9.
Example: Using three-yearly moving averages, determine the trend and short-term-error.
Year
Production in (000
1987
1988
1989
1990
1991
tonnes)
21
22
23
25
24
tonnes)
22
25
26
27
26
1992
1993
1994
1995
1996
Solution: The moving average calculation for the first 3 years is:
Moving average (year 1-3) = 21+22+23 / 3 = 22
Similarly, the moving average calculation for the next 3 years is: 22 + 23 + 25
Moving average (year 2-4) = 22+23+25/3 = 22.33
A complete summary of 3-year moving average calculations is given in Table 7.1
Table: Calculation of Trend and Short-term Fluctuations
Year (Y)
Production
Error
1987
21
9 Supra note 5
14
1988
22
66
22.00
1989
23
70
23.33
-0.33
1990
25
72
24.00
1.00
1991
24
71
23.67
0.33
1992
22
71
23.67
-1.67
1993
25
73
24.33
0.67
1994
26
78
26.00
1995
27
79
26.33
0.67
1996
26
Production
1987
1988
lbs)
464
515
(million Year
Production
1992
1993
million lbs)
540
557
(in
10 Palm, F. C., GARCH Models of Volatility: In Handbook of Statistics, Vol. 14., Ed, Amsterdam (1996) ,
209240.
15
1989
1990
1991
518
467
502
1994
1995
1996
571
586
612
Solution:
The first 4-year moving average is: MA3(4) = 464 + 515 + 518+ 467/4
= 1964/4
= 491.00
This moving average is centered on the middle value, that is, the third year of the series.
Similarly,
MA4(4)= 515+518+467+502/4
= 2002/4
= 500.50
This moving average is centered on the fourth year of the series.
Table: Presents the data along with the computations of 4-year moving averages.
Table: Calculation of Trend and Short-term Fluctuations
Year
totals
1964
2002
2027
2066
2170
2254
average
Average
491.00
500.50
506.75
516.50
542.50
563.50
Centered
495.75
503.62
511.62
529.60
553.00
572.00
1987
1988
1989
1990
1991
1992
1993
1994
464
515
518
467
502
540
557
571
1995
586
581.50
1996
612
16
Month
Last month
9 10 11 12
Actual Sales
10
12
13
16
19
23
26
17
30
28
10
18
11
16
12
14
Sales
of
1993
1994
1995
1996
(thousand units)
102
105
114
110
Firm Year
1997
1998
1999
Sales
of
Firm
(thousand units
108
116
112
12 R.P. Hooda, Statistics for Business and Economics, Universal Publishing House, (2012)
18
Solution: Since numbers of years are odd in number, therefore divide the data into equal
parts (A and B) of 3 years ignoring the middle year (1996). The average of part A and B is
y1A= 102+105+114/3 units
= 321/3 units
= 107 units
y1B = 108+116+112/3 units
= 336/2 units
= 112 units
Part A is centered upon 1994 and part B on 1998. Plot points 107 and 112 against their middle
years, 1994 and 1998.
To calculate the time series y1= a + bx, we need
Slope b = y/ x
= Change in sales/ Change in year
= 112-107/1998-1994
= 5/4
= 1.25
Intercept = a = 107 units at 1994
Thus, the trend line is:
y1 = 107 + 1.25x
Since 2002 is 8 year distant from the origin (1994), therefore we have
y1 = 107 + 1.25(8) = 117
19
20
6. CONCLUSION
Time Series is an important statistical technique to measure the various trends like the
cyclical variations, trends of time, and seasonal variations. The Moving Averages method
successfully breaks down time periods into components and then account for the factors of
different variations like time and seasonal variations. Time series data have a natural temporal
ordering. The moving averages method, weighted averages method and the semi average
method though effective just describe the values over definite and integer like periods of
time. The moving averages method cannot work when the time period and the carious
variations are non-integer or in fractions or decimals. IN the weighted averages method
random values are attributed to the variations and time periods so as to facilitate ease in
calculation. Such weightage usually changes according to how old the value is with the oldest
values getting the least weightage. In the semi-average model the plotting of the calculated
trend values is the most important function of such model. Here, the values are divided into
two different components and after calculating the arithmetic mean is plotted on the graph so
that future statistical forecasting can be done13.
This makes time series analysis distinct from cross-sectional studies, in which there
is no natural ordering of the observations (e.g. explaining people's wages by reference to their
respective education levels, where the individuals' data could be entered in any order). Time
series analysis is also distinct from spatial data analysis where the observations typically
relate to geographical locations (e.g. accounting for house prices by the location as well as
the intrinsic characteristics of the houses).
13 Gershenfeld .N, The Nature of Mathematical Modeling. New York: Cambridge University Press.
pp. 205208 (1991)
21
I.
II.
III.
IV.
BIBLIOGRAPHY
P. Bougerol and N. Picard. Stationarity of GARCH processes and some non-negative
time series. J. Econometrics, 52:115{127, 1992a.}
G. E. P. Box and G. M. Jenkins. Time Series Analysis, Forecasting and Control.
Cambridge University Press, Oakland, 1970.
D.R. Brillinger. Time Series: Data Analysis and Theory. SIAM Classics, 2001.
P. Brockwell and R. Davis. Time Series: Theory and Methods. Springer, New York,
V.
1998.
Y. Dwivedi and S. Subba Rao. A test for second order Stationarity based on the
VI.
VII.
VIII.
IX.
X.
Press (1990).
S.P.Gupta, Statistical Methods, Sultan Chand & Sons, Forty-fourth Edition, (2015)
pp.613
Thomas Holger Schreiber, Nonlinear Time Series Analysis, London, Cambridge
University Press (2004).
Brockwell, P. J. and Davis, R. A., Time Series: Theory and Methods, 2nd Edition,
Springer-Verlag, New York (1991)
22