Page 281 - DECO504_STATISTICAL_METHODS_IN_ECONOMICS_ENGLISH
P. 281

Unit 22: Time Series Analysis—Introduction and Components of Time Series


            The objectives of analysis of a time series are (l) to give a general description of the past behaviour of  Notes
            the series, (2) to analyse the past behaviour, and (3) to attempt to forecast the future behaviour on the
            basis of the past behaviour; however, great caution is needed to do this.
            A forecast does not tell what will happen but indicates what would happen if the past behaviour (as
            reflected in trends etc.) continues.
            The techniques of time series analysis have largely been developed by economists. Empirical
            investigations dealing with economic theory are largely dependent on time series analysists. Social
            scientists, in general, do not have the privilege of conducting studies through laboratory
            experimentation. Studies are to be based on time series data collected over time in such cases. For
            example, trade cycles are important to economists and others in business and commerce. The exact
            behaviour of the cycles and their causes are of interest to them. Various theories explaining the
            phenomena are put forward. Analysis of time series provides an important tool for testing the theories
            and the explanations. Consumer behaviour is studied mainly with the help of time series data.
            The analysis of time series plays an important role in empirical investigations, leading to quantitative
            revolution, in economics and in several other areas of social sciences and even in biological sciences.
            Thus political economy (usually known as economics) has been described as ‘the oldest of the arts,
            the newest of the sciences-indeed the queen of the social sciences.’
            We shall now discuss the techniques used in the analysis of time series. We begin with the main
            components or characteristic movements in a time series.




                        The analysis of time series is of interest in several areas, such as economics, commerce,
                        business, sociology, geography, meteorology, demography, public health, biology,
                        and so on.


            Objectives or Importance of Time Series Analysis
            In the words of Prof. Hirsch, “The main objective in analysing of time series is to understand, interpret and
            evaluate changes in economic phenomena in the hope of most correctly anticipating the course of future events.”
            Following are the main objectives of time series analysis.
            (1)  Study of the Past Behaviour of the Data: The purpose of time series analysis is to study the
                past behaviour of the data to easily understand what changes have taken place in the past.
            (2)  To Forecast Future Behaviour: The second objective of time series analysis is to predict the
                future behaviour of a particular variable. Time series can play an important role not in making
                short range estimates for a year or two a head but also estimating the probable seasonal variations
                within a year.
            (3)  Comparison with other Series: Time series analysis is helpful in making a comparison between
                the behaviour of different time series. We can make this comparison by knowing the causes of
                variations in two time series.
            (4)  Study of Present Fluctuations: Time series analysis is helpful in studying the present fluctuations
                in the economic variables like, national income, cost, prices, production, etc. It enables us to
                know achievements and failures regarding a particular variable.
            (5)  Estimation of Trade Cycles: The basic objective of time series analysis is to estimate the trade
                cycles. The businessmen can avoid their losses and get profits with the help of the estimation of
                trade cycles.
            22.2 Components of Time Series

            Variations in Time Series

            The term time senes are used to refer to any group of statistical information collected at regular
            intervals of time.




                                             LOVELY PROFESSIONAL UNIVERSITY                                      275
   276   277   278   279   280   281   282   283   284   285   286