RBI Governor Shaktikanta Das cautions against cherry-picking of data by experts; argues, "correlations can conveniently be extended to establish spurious causality"

Reserve Bank Goverr Shaktikanta Das on Friday cautioned against cherry-picking of data by experts, saying correlations can conveniently be used to establish "spurious causality" to achieve desired results. His remarks come in backdrop of recent controversies over GDP data after former Chief Ecomic visor Arvind Subramanian in a research paper claimed that India has over-estimated its ecomic growth.

Ecomic visory Council to Prime Minister (EAC-PM) h, however, rejected claim, saying Subramanian's analysis was based on "cherry-picking" of high-frequency indicators while igring data on services, agriculture and robust tax collection. Speaking at RBI's 13th Annual Conference, on eve of 'Statistics Day', Das said in recent times re have been animated discussions on precision of statistical methods.

RE | PM's Ecomic visory Council Publishes Point-to-point Rebuttal "debunking In Entirety" Former CEA Dr Arvind Subramanian's Claims Of GDP Overestimation. Re Here

" doctrine - even trition - of statistical significance in scientific research has come under some cloud for its veracity and journal American Statistician published a special edition on this issue earlier this year," he said.

Das also said that critics allege that se tests are susceptible to manipulation in order to make desired results significant, and undesired results n-significant. Furr, some important results may be discarded at conception level itself just because y are highly unlikely, he said.

"Similarly, an opportunity to cherry-pick variables is available. In or words, correlations can conveniently be extended to establish spurious causality," Goverr said.

RE: Bank fraud touches unprecedented Rs 71,500 crore in 2018-19: Reserve Bank

central bank will continue to refine methodologies and use cutting edge techniques for forecasting of macroecomic developments on an ongoing basis. RBI held its 13th Annual Conference with me 'Statistics for Future - Select Issues', on eve of Statistics Day' -- designated by Government of India on June 29 to commemorate birth anniversary of rewned statistician Prasanta Chandra Mahalabis. In his inaugural dress, Das said in an era of internet and social media, re is substitute for rigorous statistical testing for establishing empirical regularities. In a deluge of data and results, what is vital is t just kwing which facts warrant importance, but also which are to be igred, or even strongly refuted, he said.

"In Reserve Bank, we will continue to refine methodologies used for forecasting and assessment of macroecomic developments on an ongoing basis."

"Research and analytics using cutting edge techniques will be pursued and in particular, wcasting of growth and inflation will be furr strengned," Goverr said.

He furr said big data is new buzzword in world of statistics, and it has alrey started changing way world views itself. Corporations, he said, are making large investments to predict behavior of consumers by exploiting vances me in field of data analytics. This information techlogy revolution has also created problems of plenty, underscoring need for rigorous processes of classification, aggregation and analysis.

"Big data analytics are being increasingly employed to assess food inflation, to develop risk profiles and stress scenarios for corporate sector and to conduct sentiment analysis with artificial intelligence and machine learning techniques," Das said.

Mahalabis, member of first Planning Commission of India, is remembered for founding Indian Statistical Institute (ISI); for his path-breaking work on nationwide sample surveys; and for providing architecture of second five year plan (FYP) model of rapid industrialization.