The Third Statistics Day of India

S Kunjabihari *

India has been observing the 29th of June, the birth anniversary of the late Prof Prasanta Chandra Mahalanobis, as the National Statistics Day, effective from the 29th June of the year 2007, and the 114th birth anniversary of Prof P C Mahalanobis.

This is in recognition of the vast contribution made by him in the field of statistics, in particular in the realm of planning for economic and social development. Prof Mahalanobis pioneered almost singularly, the setting up of Indian Statistical Institute, in the then Calcutta, largely to initiate adoption of statistical methodologies, in solving social and economic problems, furthered the cause of critical investigation, through rigorous innovative technique and analysis of issues in quantitative economics.

He succeeded in collecting a galaxy of young and great minds to painstaking research on the premise that a sound decision-making in the society, for the many economic problems, should be based on objective statistical evidence, not otherwise. He is regarded as one who had paved the foundations of the modern statistical system in the country.

Dr Manmohan, PM of India, observed on 20, June, 2007, "Our Government resolved last year on the occasion of the Platinum Jubilee of the Indian Statistical Institute (ISI), to honor the memory of one of our greatest statistician and development planners, Professor Prasanta Chandra Mahalanobis, by declaring his birth anniversary, the 29th June as Statistics Day.

The year 1931 will ever remain a red letter day in the annals of development of a firm footing of Statistics. In the precincts of the Physics Department of the Presidency College, a make-shift Statistical Laboratory instituted by the Prof and a handful of the best minds in the fields of Mathematics and Physics of those days sowed the seed of modern statistics.

They eventually founded the Indian Statistical Institute (SI), on 17 Dec, 1931. In 1959, by an act of the Parliament, as an Institute of National Importance and empowered to award degrees in Statistics like a University. P C Mahalanobis is best described, by Prof CR Rao, who succeeded him as Director of the ISI (Calcutta), "� a physicist by training, a statistician by instinct and an economist by conviction".

R A Fisher, one of the greatest mathematical statistician of the time, observed, "The ISI has taken the lead in the original development of the technique of sample surveys, the most potential fact finding process available to the administrator."

Why Statistics: The indispensability of basic statistics for formulating and implementing plans of economic and social development in the countries, the world over, developed or underdeveloped hardly needs any emphasis. Availability of data alone is not adequate; these should be available timely, should be reliable, and be able to stand the test of time; authentic and comprehensive with an eye to maneuverability for one to one correspondence.

To this end in view, apart from efforts of individual countries, the United Nationals had been pioneering the methodologies for generating basic statistical series, largely for the consumption of the less developed countries. Such bodies like the Asian Statisticians Commission, meet regularly for improving, not only the availability but also the methodologies for collection, the quality and content of the series.

Such collective and unified measures are useful for standardizing the concept, methodologies, varieties, frequencies etc for comparison at the international level. Apart from consistency and uniformity, an integrated system is attempted. The requirement of a host of data, for instance, for compiling national accounts could be enormous.

Considering the enormously of the effort in collecting, compiling, collating various statistics, there is also the emerging imperative for identifying scale of priorities of collecting and compiling these series. Galileo Philosophized, "Measure. measure, measure. Measure again and again to find out the difference and the difference of the difference."

Obviously, different sets of data would be called for different functions and objectives. For instance, an inventory of the economy would require basic economic and social conditions and available resources and the use to which these are priotised. Such an inventory would be required for setting meaningful and realistic targets for economic and social changes.

Such targets are useful for, inter-a-lia, character and level of production, living conditions, supply and demand of goods and services, trades, sources of external and internal finances etc. It would thus, be apparent that data are required on changes in economic and social conditions. P C Mahalanobis observed, "Statistics is the only court of appeal to new knowledge; it is the universal tool of inductive inferences in natural and social sciences, and technological applications".

An operational plan must be necessarily quantitative in nature. Parameters like-rate of increase of income, percentage consumption and savings, investments efforts, rate of growth of the economy, structure of trade, taxation, prices etc are needed for realistic policy formulation. The extent to which data will reveal discrepancies, between planned and actual magnitudes will depend upon availability of these data.

In other words, the refiner and the more reliable, the series are, the quality of the projections would be. RA Fisher, a Statistician of international repute in the post-independence era, and a contemporary of the late, PC Mahalanobis, was very candid about the poor quality and variety of data for any exercise.

He was emphatic that Statisticians are not Alchemists, who can turn out gold out of mud; but that the quality of the output is as good as that of the input. The quality of basic statistics, and methodologies developed, in those years; in India were one of the best. The words of H Hotelling, could be recalled, "No technique of random sample has, so far as I can find, been developed in the United States or elsewhere, which can compare in accuracy with that described by Prof Mahalanobis.

Why Price Statistics: In the realm of "Price Statistics," the theme-focus of, this year's Statistics Day, we may recall the relentless research towards quest for quality in areas like, average annual income, consumption, expenditure, savings quantity and value of foodstuffs etc. These would need be further broken down to, say, distribution of households and income and expenditures units by pattern of annual expenditure and savings which would lead to such parameters like �Index Number of Consumer Prices that too on a regular basis, say, monthly, yearly etc.

Such studies would be achievable through sample surveys which again in turn, be based on a more wieldy data, say, census figures. The intricacies are sizable, indeed. For instance, CPI, a measure of the average change over time in the prices paid by consumers also reflects the expenditure patterns.

The CPI market basket is developed from detailed expenditure information provided by families and individuals on what they actually bought. The consumer expenditure pattern comes into play. In advanced countries, the sample families maintain family diaries listing everything they bought. In the US, for such statistics, the expenditure information came from around 28,000 weekly diaries and 60,000 quarterly interviews. The prices of about 80,000 items every month are recorded for a sample of consumers.

The changes in prices over a period of time can be studied with the statistical device of Index Numbers of prices; Wholesale Price Index (WPI), for instance, helps in understanding fluctuations of prices relating to bulk transactions. A number of Government Of India departments, in particular, the Planning Commission, the Central Statistical Organisation, the Reserve Bank of India etc base their schemes and to modify their monetary and fiscal policies in the light of the price trends.

Price statistics would be required for framing pricing policy, trading policy, fiscal and monitory policy, and foreign exchange policy, coefficients of percentage of unit price index of export to unit price index of imports; income elasticity of import and export; propensity of import, parity of purchasing power of currency to mention a few.

In a country of India's geographical vastness, coupled with the topographical disparities with acute inadequacy of transport and communication infrastructures, regional disparities in prices are on considerable concern in adopting a unified index. There would therefore be anxiety regarding possible similarity or otherwise of price movements in remote regions, and whether there could be any casual relationship between movements of prices and other economic variables, the imperatives of the observed price fluctuations.

For instance, agricultural prices are subject to significant seasonal and even regional fluctuations, and this is one sector of the economy which contributes around 50% of the National or the State income and therefore there may be aberrations in the price index unless these factors are taken care of. An authentic price index is handy for the Indian cultivators are responsive to the price changes of commodities.

It would be appropriate to recall, an observation of Edmund Burke, "The age of chivalry is gone; that of sophisters, economists and calculators has succeeded". Indeed, there has to be, all the more need for constant refinement in the process of collecting, collating, calculating indices of various parameters of the economy, more than any time before, towards the universal goal of a more prosperous social order.