Prasanta Chandra Mahalanobis

Prasanta Chandra Mahalanobis

P. C. Mahalanobis (Madan)
Born Bengali: প্রশান্ত চন্দ্র মহলানবিস
29 June 1893
Calcutta, Bengal, British India
Died 28 June 1972 (aged 78)
Calcutta, West Bengal, India
Residence India, United Kingdom, United States
Nationality India
Fields Mathematics, Statistics
Institutions University of Cambridge
Indian Statistical Institute
Alma mater University of Calcutta
King's College, Cambridge[1]
Doctoral advisor W. H. Macaulay[2]
Doctoral students Samarendra Roy[2]
Known for Mahalanobis distance
Notable awards Padma Vibhushan (1968)
Fellow of the Royal Society[1]
Weldon Memorial Prize

Signature

Prasanta Chandra Mahalanobis FRS[1] (Bengali: প্রশান্ত চন্দ্র মহলানবিস) (29 June 1893 – 28 June 1972) was an Indian scientist and applied statistician. He is best remembered for the Mahalanobis distance, a statistical measure. He made pioneering studies in anthropometry in India. He founded the Indian Statistical Institute, and contributed to the design of large-scale sample surveys.[1][3][4][5]

Early life

Mahalanobis belonged to a family of Bengali landed gentry 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 Prize–winning 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 center of the Brahmo Samaj. Gurucharan married a widow against social traditions. His elder son Subodhchandra (1867–1953) was a distinguished educationist who studied physiology at Edinburgh University. He was a Fellow of the Royal Society. He was the Head of the Dept. of Physiology, University of Cardiff (first Indian to occupy this post in a British university). In 1900, he founded Dept. of Physiology in the Presidency College, Calcutta. Subodhchandra also became a member of the Senate of the Calcutta University. Gurucharan's younger son, Prabodh Chandra (1869-1942) was the father of P. C. Mahalanobis. 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 graduating in 1908. He then joined the Presidency College, Calcutta where he was taught by teachers who included Jagadish Chandra Bose, Sarada Prasanna Das and Prafulla Chandra Ray. Meghnad Saha was a year junior and Subhas Chandra Bose was two years his junior at college.[6] and received a Bachelor of Science degree with honours in physics in 1912. He left for England in 1913 to join the University of London. He however missed a train and stayed with a friend at King's College, Cambridge. He was impressed by King's College Chapel there and his host's friend M. A. Candeth suggested that he could try joining there, which he did. He did well in his studies at King's, but also took an interest in cross-country walking and punting on the river. He interacted with the mathematical genius Srinivasa Ramanujan 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. They married on 27 February 1923 although her father did not completely approve of it. The contention was partly due to Mahalanobis's opposition to 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's maternal uncle took part in the wedding ceremony in place of the father of the bride.[1]

Indian Statistical Institute

Mahalanobis memorial at ISI Delhi.

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 17 December 1931 with Pramatha Nath Banerji (Minto Professor of Economics), Nikhil Ranjan Sen (Khaira Professor of Applied Mathematics) and Sir R. N. Mukherji. The 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, Gopinath Kallianpur, D. B. Lahiri and C. R. Rao. 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 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 United States 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 the ISI due to frustrations with the administration and disagreements with Mahalanobis's policies. He was also very concerned with the frequent travels and absence of the director and complained that the ... journeyings of our Director define a novel random vector. Haldane however helped the ISI grow in biometrics.[7]

In 1959, the institute was declared as an institute of national importance and a deemed university.[1]

Contributions to statistics

Mahalanobis distance

Main article: 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 Annandale asking him to analyse anthropometric measurements of Anglo-Indians in Calcutta. Mahalanobis had been influenced by the anthropometric studies published in the journal Biometrika and he chose to ask the questions on what factors influence the formation of European and Indian marriages. He wanted to examine if the Indian side came from any specific castes. He used the data collected by Annandale and the caste specific measurements made by Herbert Risley to come up with the conclusion that the sample represented a mix of Europeans mainly with people from Bengal and Punjab but not with those from the Northwest Frontier Provinces or from Chhota Nagpur. He also concluded that the intermixture more frequently involved the higher castes than the lower ones.[8][9] This analysis was described by his first scientific paper in 1922.[10] During the course of these studies he found a way of comparing and grouping populations using a multivariate distance measure. This measure, denoted "D2" and now eponymously named 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 analysing university exam results, anthropometric measurements on Anglo-Indians of Calcutta and some meteorological 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 lifelong friendship. He also worked on schemes to prevent floods.

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 ISI 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.[11][12][13]

Linguistics

Mahalanobis also started research in the field of quantitative linguistics and language planning in the Linguistic Research Unit of the Indian Statistical Institute.[14] He also worked on Speech Pathology in collaboration with Djordge Kostic, Rhea Das and Alakananda Mitter and made some contributions to the field of language correction.[15]

Later life

In later life, Mahalanobis was a member of the planning commission[16] contributed prominently to newly independent India's five-year plans starting from the second. In the second five-year plan he emphasised industrialisation 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 industrialisation 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 deindustrialisation in India and correct some previous census methodology errors and entrusted this project to Daniel Thorner.[17]

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.

Honours

The government of India decided in 2006 to celebrate his birthday, 29 June, as National Statistical Day.[19][20]

See also

References

  1. 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 1.12 1.13 1.14 1.15 Rao, C. R. (1973). "Prasantha Chandra Mahalanobis 1893-1972". Biographical Memoirs of Fellows of the Royal Society 19: 454. doi:10.1098/rsbm.1973.0017.
  2. 2.0 2.1 Prasanta Chandra Mahalanobis at the Mathematics Genealogy Project
  3. Hagger-Johnson, G. (2005). "Mahalanobis, Prasanta Chandra". Encyclopedia of Statistics in Behavioral Science. doi:10.1002/0470013192.bsa360. ISBN 0470860804.
  4. O'Connor, John J.; Robertson, Edmund F., "Prasanta Chandra Mahalanobis", MacTutor History of Mathematics archive, University of St Andrews.
  5. Ghosh, J. K.; Majumder, P. P. (2005). "Mahalanobis, Prasanta Chandra". Encyclopedia of Biostatistics. doi:10.1002/0470011815.b2a17090. ISBN 047084907X.
  6. Venkataraman, G. (1995). Saha and his formula. Hyderabad: Universities Press. p. 3.
  7. Dronamraju, K. 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 41 (2): 211–237. doi:10.1098/rsnr.1987.0006. PMID 11622022.
  8. Dasgupta, Somesh (1993). "The evolution of the D2-statistic of Mahalanobis". Sankhya 55: 442–459.
  9. Mahalanobis, P C (1927). "Analysis of race-mixture in Bengal". J. Proc. Asiatic Soc. Bengal 23.
  10. Mahalanobis, PC (1922). "Anthropological observations on the Anglo-Indians of Calcutta. Part I. Analysis of male stature.". Records of the Indian Museum 23: 1–96.
  11. 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
  12. 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.
  13. 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. doi:10.2307/1403563. JSTOR 1403563.
  14. Debnath, Sukesh; Debaprasad Bandyopadhyay (1996). "Mahalanobis as a Language Planner". Indian Journal of Applied Linguistics XXII (1): 49–57.
  15. "Colti Bhasar Banan". Prabasi. 1925.
  16. The Hindu dated 15th May, 2003
  17. Das, Gurucharan. 2000 India Unbound: The Social and Economic Revolution from Independence to the Global Information Age Anchor Books. pp. 432 ISBN 0-375-41164-X
  18. Royal Society citation
  19. The Statesman 25 December 2006
  20. 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

Bibliography

External links