Prasanta Chandra Mahalanobis
From Wikipedia, the free encyclopedia
Prasanta Chandra Mahalanobis (Bangla: প্রশান্ত চন্দ্র মহলানবিস) (June 29, 1893–June 28, 1972) was an Indian scientist and applied statistician. He is best remembered for the Mahalanobis distance, a statistical measure. He did pioneering work on anthropometric variation in India. He founded the Indian Statistical Institute, and contributed to large scale sample surveys.[1][2]
Contents |
[edit] Early life
Mahalanobis belonged to a family of landed Bengali aristocracy who lived in Bikrampur (now in Bangladesh). His grandfather Gurucharan (1833-1916) moved to Calcutta in 1854 and built up a business, starting a chemist shop in 1860. Gurucharan was influenced by Debendranath Tagore (1817-1905), father of the Nobel poet, Rabindranath Tagore. Gurucharan was actively involved in social movements such as the Brahmo Samaj, acting as its Treasurer and President. His house on 210 Cornwallis Street was the centre of the Brahmo Samaj. Gurucharan married a widow against social traditions. His elder son Subodhchandra (1867-1954) was the father of P. C. Mahalanobis. He was a distinguished educationist who studied physiology at Edinburgh University and later became a Professor at the Presidency College to become head of the department of Physiology. Subodhchandra also became a member of the Senate of the Calcutta University. Born in the house at 210 Cornwallis Street, P. C. Mahalanobis, grew up in a socially active family surrounded by intellectuals and reformers.[1]
Mahalanobis received his early schooling at the Brahmo Boys School in Calcutta passng in 1908. He then joined the Presidency College, Calcutta and received a B.Sc. degree with honours in physics in 1912. He left for England in 1913 to join Cambridge. He however missed a train and stayed with a friend at King's College. He was impressed by the Chapel there and his host's friend M. A. Candeth suggested that he could try joining there and this he did. He did well in his academics, but also took an interest in cross-country walking and punting on the river. He interacted with the mathematical genius Srinivasa Ramanujam during the latter's time at Cambridge. After his Tripos in physics, Mahalanobis worked with C. T. R. Wilson at the Cavendish Laboratory. He took a short break and went to India and here he was introduced to the Principal of Presidency College and was invited to take classes in physics.[1]
He went back to England and was introduced to the journal Biometrika. This interested him so much that he bought a complete set and took them to India. He discovered the utility of statistics to problems in meteorology, anthropology and began working on it on his journey back to India.[1]
In Calcutta, Mahalanobis met Nirmalkumari, daughter of Herambhachandra Maitra, a leading educationist and member of the Brahmo Samaj. He married her on 27 February 1923 although her father did not completely approve of it. The contention was partly due to Mahalanobis' opposition of various clauses in the membership of the student wing of the Brahmo samaj, including restraining members from drinking and smoking. Sir Nilratan Sircar, P. C. Mahalanobis' uncle took part in the wedding ceremony in place of the father of the bride.[1]
[edit] The Indian Statistical Institute
Many colleagues of Mahalanobis took an interest in statistics and the group grew in the Statistical Laboratory located in his room at the Presidency College, Calcutta. A meeting was called on the 17 December 1931 with Pramatha Nath Banerji (Minto Professor of Economics), Nikhil Ranjan Sen (Khaira Professor of Applied Mathematics) and Sir R. N. Mukherji. Thie meeting led to the establishment of the Indian Statistical Institute (ISI), and formally registered on 28 April 1932 as a non-profit distributing learned society under the Societies Registration Act XXI of 1860.[1]
The Institute was initially in the Physics Department of the Presidency College and the expenditure in the first year was Rs. 238. It gradually grew with the pioneering work of a group of his colleagues including S. S. Bose, J. M. Sengupta, R. C. Bose, S. N. Roy, K. R. Nair, R. R. Bahadur, G. Kallianpur and D. B. Lahiri. The institute also gained major assistance through Pitamber Pant, who was a secretary to the Prime Minister Jawaharlal Nehru. Pant was trained in statistics at the Institute and took a keen interest in the institute.[1]
In 1933, the journal Sankhya was also founded along the lines of Karl Pearson's Biometrika.[1]
The Institute started a training section in 1938. Many of the early workers left the ISI for careers in the USA and with the government of India. Mahalanobis invited J. B. S. Haldane to join him at the ISI and Haldane joined as a Research Professor from August 1957 and stayed on until February 1961. He resigned from ISI due to frustrations with the administration and disagreements with Mahalanobis's administrative policies. He was also very concerned with the frequent travels and absence of the director and wrote The journeyings of our Director define a novel random vector. Haldane however helped the ISI grow in biometrics.[3]
In 1959 the Institute was declared as an Institute of national importance and a deemed university.[1]
[edit] Contributions to statistics
[edit] Mahalanobis distance
A chance meeting with Nelson Annandale, then the director of the Zoological Survey of India, at the 1920 Nagpur session of the Indian Science Congress led to a problem in anthropology. Annandale asked him to analyse anthropometric measurements of Anglo-Indians in Calcutta and this led to his first scientific paper in 1922. During the course of these studies he found a way of comparing and grouping populations using a multivariate distance measure. This measure, D2, which is now named after him as Mahalanobis distance, is independent of measurement scale.[1]
Inspired by Biometrika and mentored by Acharya Brajendra Nath Seal he started his statistical work. Initially he worked on analyzing university exam results, anthropometric measurements on Anglo-Indians of Calcutta and some metrological problems. He also worked as a meteorologist for some time. In 1924, when he was working on the probable error of results of agricultural experiments, he met Ronald Fisher, with whom he established a life-long friendship. He also worked on schemes to prevent floods.
[edit] Sample surveys
His most important contributions are related to large scale sample surveys. He introduced the concept of pilot surveys and advocated the usefulness of sampling methods. Early surveys began between 1937 to 1944 and included topics such as consumer expenditure, tea-drinking habits, public opinion, crop acreage and plant disease. Harold Hotelling wrote: "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 Professor Mahalanobis" and Sir R. A. Fisher commented that "The I.S.I. has taken the lead in the original development of the technique of sample surveys, the most potent fact finding process available to the administration".[1]
He introduced a method for estimating crop yields which involved statisticians sampling in the fields by cutting crops in a circle of diameter 4 feet. Others such as P. V. Sukhatme and V. G. Panse who began to work on crop surveys with the Indian Council of Agricultural Research and the Indian Agricultural Statistics Research Institute suggested that a survey system should make use of the existing administrative framework. The differences in opinion led to acrimony and there was little interaction between Mahalanobis and agricultural research in later years.[4][5][6]
In later life, Mahalanobis was a member of the planning commission[7] contributed prominently to newly independent India's five-year plans starting from the second. In the second five-year plan he emphasised industrialization on the basis of a two-sector model.[1] His variant of Wassily Leontief's Input-output model, the Mahalanobis model, was employed in the Second Five Year Plan, which worked towards the rapid industrialization of India and with other colleagues at his institute, he played a key role in the development of a statistical infrastructure. He encouraged a project to assess de-industrialization in India and correct some previous census methodology errors and entrusted this project to Daniel Thorner.[8]
Mahalanobis also had an abiding interest in cultural pursuits and served as secretary to Rabindranath Tagore, particularly during the latter's foreign travels, and also worked at his Visva-Bharati University, for some time. He received one of the highest civilian awards, the Padma Vibhushan from the Government of India for his contribution to science and services to the country.
Mahalanobis died on 28 June, 1972, a day before his seventy-ninth birthday. Even at this age, he was still active doing research work and discharging his duties as the Secretary and Director of the Indian Statistical Institute and as the Honorary Statistical Advisor to the Cabinet of the Government of India.
[edit] Honours
- Weldon Medal from Oxford University (1944)
- Fellow of the Royal Society, London (1945)
- President of Indian Science Congress (1950)
- Fellow of the Econometric Society, U.S.A. (1951)
- Fellow of the Pakistan Statistical Association (1952)
- Honorary Fellow of the Royal Statistical Society, U.K. (1954)
- Sir Deviprasad Sarvadhikari Gold Medal (1957)
- Foreign member of the Soviet Academy of Sciences (1958)
- Honorary Fellow of King's College Cambridge (1959)
- Fellow of the American Statistical Association (1961)
- Durgaprasad Khaitan Gold Medal (1961)
- Padma Vibhushan (1968)
- Srinivasa Ramanujam Gold Medal (1968)
The government of India decided in 2006 to celebrate his birthday, 29 June, as National Statistical Day.[9][10]
[edit] References
- ^ a b c d e f g h i j k l Rao, C. R. (1973) Prasantha Chandra Mahalanobis. 1893-1972. Biographical Memoirs of Fellows of the Royal Society. 19:454-492
- ^ Rudra, A. (1996), Prasanta Chandra Mahalanobis: A Biography. Oxford University Press.
- ^ Dronamraju, Krishna R. (1987). "On Some Aspects of the Life and Work of John Burdon Sanderson Haldane, F.R.S., in India". Notes and Records of the Royal Society of London 41 (2): 211-237.
- ^ Rao, J.N.K. (2006) Interplay Between Sample Survey Theory and Practice: An Appraisal. Survey Methodology Vol. 31, No. 2, pp. 117-138. Statistics Canada, Catalogue No. 12-001 PDF
- ^ Adhikari, B.P (1990). Social construction of the statistical estimation of crop yield. Paper presented at the XII World Congress of Sociology of the Internutionul Sociologicul Associution, Madrid, Spain.
- ^ Ghosh, J.K.; P. Maiti; T. J. Rao; B. K. Sinha (1999). "Evolution of Statistics in India". Revue Internationale de Statistique 67 (1): 13-34.
- ^ The Hindu dated 15th May, 2003
- ^ Das, Gurucharan. 2000 India Unbound: The Social and Economic Revolution from Independence to the Global Information Age Anchor Books. pp. 432 ISBN 037541164X
- ^ The Statesman 25 December 2006
- ^ Mohan, Rakesh 2007 Statistical system of India – some reflections. Reserve Bank of India, Department of Statistical Analysis and Computer Services, Mumbai, 29 June 2007. PDF
