Page 334 - Special Topic Session (STS) - Volume 4

STS700 Suman R. A.

Accommodating Big Data in Nepalese Statistical
System: Challenges and opportunities
Suman Raj Aryal
Cetral Bureau of Statistics

Abstract
Intellectual discourse on big data has already been introduced in the national
statistical system (NSS) of Nepal. Nevertheless, the discourse in not expanded
to the wide range of data producers and users of data. The big data is not yet
officially used as an alternative source of data for decision making. Some
private organizations working in the field of data are also found enthusiastic
on big data. The recent National Strategy for the Development of Statistics
(NSDS), 2018-2023 of Nepal has explicitly recognized the big data as a new
initiative which sooner or later can be a complimentary source of data. As such,
big data initiative is expected to stand as additional means of data supply
regardless of its complexity and statistical aspects.
Irrespective of traditional sources of data, viz., census, surveys, administrative
data and research outputs the big data by its very nature are difficult to handle
in a country like Nepal where the statistical infrastructure is not built
adequately. It needs, on the one hand, huge amount of investment to mine
and process the big data and a legal acceptance from the State on the other.
The Fundamental Principles of Official Statistics (FPOS) should not be violated
while recognizing the big data. The question of confidentiality and the data
security are equally important. Neither the national statistics offices (NSOs)
nor any other agencies in the NSS have capacity to process, store and
disseminate the big data at the moment. There is another important question
which is about the selection of sources of big data. In the very beginning of
this initiative, Nepal is not in a position to mine all sources of big data. The big
data generated from social media are less likely to be appropriated since these
require approval from multinational private companies. Again, the FPOS puts
a dilemma, i.e., whether or not data generated by the business institutions
should be regarded as official statistics. Validation of big data should pass
through several statistical gates. At the moment, Nepal can use the CDR data
for various purposes. The call detail records (CDR) and ATM data can be used
to assess the migration, access to public services, viz, mobile banking, online
shopping, unauthorized trade in the open border, employment mobility,
weather, land use and so on by real time.
Big data has, as in other countries, provided us an opportunity to bring
together the statisticians, the data scientists and the IT professionals. The
knowledge and expertise of these three core professionals are very likely to

323 | I S I W S C 2 0 1 9