affected by the advertising budget, the media plan, the content of the advertisements, number of salesmen, price of the product, efficiency of the distribution network and a host of other variables. For determining causal relationship involving two or more variables, multi-variate statistical techniques are applicable. The most important of these are the multiple regression analysis, discriminant analysis and factor analysis. v Time Series Analysis A time series consists of a set of data arranged in some desired manner recorded either at successive points in time or over successive periods of time. The changes in such type of data from time to time are considered as the resultant of the combined impact of a force that is constantly at work. This force has four components i Editing time series data, ii secular trend, iii periodic changes, cyclical changes and seasonal variations, and iv irregular or random variations. With time series analysis, you can isolate and measure the separate effects of these forces on the variables. Examples of these changes can be seen, if you start measuring increase in cost of living, increase of population over a period of time, growth of agricultural food production in India over the last fifteen years, seasonal requirement of items, impact of floods, strikes, wars and so on. vii Index Numbers Index number is a relative number that is used to represent the net result of change in a group of related variables that has some over a period of time. Index numbers are stated in the form of percentages. For example, if we say that the index of prices is 105, it means that prices have gone up by 5 as compared to a point of reference, called the base year. If the prices of the year 1985 are compared with those of 1975, the year 1985 would be called "given or current year" and the year 1975 would be termed as the "base year". Index numbers are also used in comparing production, sales price, volume employment, etc. changes over period of time, relative to a base. viii Sampling and Statistical Inference In many cases due to shortage of time, cost or non-availability of data, only limited part or section of the universe or population is examined to i get information about the universe as clearly and precisely as possible, and ii determine the reliability of the estimates. This small part or section selected from the universe is called the sample, and1the process of selections such a section or past is called sampling Scheme of drawing samples from the population can be classified .into two broad categories a Random sampling schemes In these schemes drawing of elements from the population is random and selection of an element is made in such a way that every element has equal chance probability of being selected. b Non-random sampling schemes In these schemes, drawing of elements from the population is based on the choice or purpose of selector. The sampling analysis through the use of various tests' namely Z-normal distribution, student's t' distribution F-distribution and x2-distribution make possible to derive inferences about population parameters with specified level of significance and given degree of freedom. You will read about a number of tests in this block to derive inference about population parameters.
ADVANTAGES OF QUANTITATIVE APPROACH