Predicting HDB Resale Prices - Conducting Linear Regression Analysis With Orange
ONS Economic Forum, Cambridge
1. Statistics for the public good
Ed Palmer
Deputy Chief Economist, Economic Advice and Analysis
Office for National Statistics
… and Clare (1991)
economic.advice@ons.gov.uk
Economic Forum, University of Cambridge
30 November 2018
2. To discuss today
• UK Statistics Authority and the Office for National
Statistics: our status and role
• What the ONS does
• Why we are transforming
• Some examples of how we are transforming
2
3. The status and role of the UK Statistics
Authority
• An independent statutory body
• Operating at arm’s length from government as a non-
ministerial department, reporting directly to the UK’s
Parliaments and Assemblies
• In law (the Statistics and Registration Services Act 2007)
our objective is:
“promoting and safeguarding the production and
publication of official statistics that serve the public
good”
• And that public good includes:
• informing the public about social and environmental matters
• assisting in the development and evaluation of public policy
• regulating quality and publicly challenging the misuse of
statistics 3
5. Who are we?
• Economic, public policy and population statistics
• Other government Departments also provide statistics,
e.g. energy, health, environment
• Our responsibility is for coherence of system as a whole
• Census provider in England and Wales
• Office locations in Newport, Titchfield, London
UK Statistics
Authority
Other
government
stats producers
Office for
National
Statistics
Office for
Statistics
Regulation
5
7. And a lot more:
GDP growth Inflation (Un)employment Wages
Trade Public finances
Regional and
country economic
data
Crime
Births, deaths and
marriages
Population Migration
Health and social
care
Personal income
and wealth
Well-being Environment …and more
7
8. Economic Statistics: transformation
8
• Users’ needs are changing
• Economy is evolving quickly
• Opportunities from new data
sources
• ONS needs to be more open
Bean
Independent Review of
UK Economic Statistics
2016
Johnson
• UK Consumer Prices
Stats – A Review 2015
Barker / Ridgeway
• National Statistics Quality
Review 2014
Best Practice
• European System of
Accounts 2010
9. Data are central to the
decisions which affect
our lives
Data are now available
from previously
unimaginable sources
We treat personal data
confidentially and make
sense of numbers for
the public good
Economic Statistics: transformation
9
10. Three ways of calculating GDP
Expenditure
GDP(E)
How
much is
spent
Output
GDP (O)
How
much is
produced
Income
GDP (I)
How
much is
earned
10
11. Three ways of calculating GDP
Expenditure
GDP(E)
How much is
spent
Use of credit card
data
Retail sales
Output
GDP (O)
How much is
produced
Use of VAT data
Purchases
Survey
Improving
Deflators
Income
GDP (I)
How much is
earned
Use of PAYE &
Self-Assessment
data
Improving the Supply-Use system 11
12. VAT: the benefits
Current system
Turnover is used as a
proxy for gross value
added
Surveys sent to 45,000
firms each month
Limited regional and local
estimates
Refreshing a survey
takes time to get new
data
Future system
VAT could give us GVA –
in time…
2m returns each quarter
Significant increase in
granularity of estimates
Admin data comes with
historical data for rapid
use
12
14. PAYE: the benefits
Current system
Average Weekly
Earnings x employment =
Salaries
Labour Force Survey
sent to 40,000
households
Monthly Wages &
Salaries Survey sent to
9,000 firms monthly
Future system
PAYE = Salaries + some
pensions
65m records within 2
weeks, including 18m
pension payments
14
15. Economic Statistics: analysis and research
Economic Statistics Centre of
Excellence
• National Accounts and Beyond GDP
• Productivity and the modern economy
• Regional and labour market statistics
• www.escoe.ac.uk
15
ONS Economic Statistics and Analysis
Strategy
• Sets out for users, stakeholders and
researchers clarity on how we are
working to improve UK economic
statistics.
• https://www.ons.gov.uk/methodology/clas
sificationsandstandards/economicstatistic
sclassifications/onseconomicstatisticsand
analysisstrategyfinancialyearending2019
16. Understanding the UK economy
Ellys Monahan
Office for National Statistics
economic.advice@ons.gov.uk
University of Cambridge
30 November 2018
17. Latest data from the ONS
• GDP
• Economic well-being
• Inflation
• Employment
• Productivity
17
18. GDP growth: 0.6% in Quarter 3 2018
18
Gross domestic product growth rate, quarter-on-quarter and quarter on same quarter a year ago
UK, Quarter 1 (Jan to Mar) 2008 to Quarter 3 (July to Sept) 2018
Source: ONS
22. Latest data from ONS
• GDP
• Economic well-being
• Inflation
• Employment
• Productivity
22
23. Measures of inflation: input producer prices,
output producer prices, consumer prices
23Source: ONS
12-month growth rates for input producer prices and output producer prices (left-hand side),
and Consumer Prices Index including owner occupiers’ housing costs (CPIH) (right-hand side)
24. What is driving the growth in consumer
prices?
24Source: ONS
Contributions to the CPIH 12-month rate: October 2016 to October 2018
25. House prices, inflation and wages
25Source: ONS, Land Registry
Index, January 2007 = 100
60
80
100
120
140
160
180
2007
JAN
2007
JUL
2008
JAN
2008
JUL
2009
JAN
2009
JUL
2010
JAN
2010
JUL
2011
JAN
2011
JUL
2012
JAN
2012
JUL
2013
JAN
2013
JUL
2014
JAN
2014
JUL
2015
JAN
2015
JUL
2016
JAN
2016
JUL
2017
JAN
2017
JUL
2018
JAN
2018
JUL
East of England Great Britain Cambridge CPIH AWE
26. Latest data from ONS
• GDP
• Economic well-being
• Inflation
• Employment
• Productivity
26
27. Labour market headline figures
27Source: ONS
UK, seasonally adjusted, Jan to Mar 2006 to July to Sep 2018
0
1
2
3
4
5
6
7
8
9
10
67
68
69
70
71
72
73
74
75
76
77
Jan-Mar
2006
Jan-Mar
2007
Jan-Mar
2008
Jan-Mar
2009
Jan-Mar
2010
Jan-Mar
2011
Jan-Mar
2012
Jan-Mar
2013
Jan-Mar
2014
Jan-Mar
2015
Jan-Mar
2016
Jan-Mar
2017
Jan-Mar
2018
Employment rate % (Age 16-64) LHS Unemployment rate % (Age 16-64) RHS
28. Latest data from ONS
• GDP
• Economic well-being
• Inflation
• Employment
• Productivity
28
29. Productivity performance is a key weakness
29Source: ONS
Output per hour and output per worker. Seasonally adjusted
Quarter 1 (Jan to Mar) 1994 to Quarter 2 (Apr to Jun) 2018, UK
30. Latest data from ONS
• GDP
• Economic well-being
• Inflation
• Employment
• Productivity
30
31. Recent developments in ONS
economic statistics - a regional
perspective
Richard Prothero
Head of ONS Centre for Subnational Analysis,
32. Devolution Programme
A programme to improve ONS regional and local statistics.
Regional Balanced GVA
Regional & Sub-Regional Household Final Consumption Expenditure
Regional Short Term Indicators
Country and Regional Public Sector Finances
Exports of Services Data
Productivity
Small Area Data
Flexible Geographies
Investigating Uses of Administrative Data
Regional Prices
Stakeholder Engagement
33. Economic growth by region/country
UK UK
-1
0
1
2
3
4
5
NorthEast
NorthWest
YorkshireandTheHumber
EastMidlands
WestMidlands
EastofEngland
London
SouthEast
SouthWest
Wales
Scotland
NorthernIreland
NorthEast
NorthWest
YorkshireandTheHumber
EastMidlands
WestMidlands
EastofEngland
London
SouthEast
SouthWest
Wales
Scotland
NorthernIreland
1998-2007 2010-2016
%
Annual growth rates of real GVA by NUTS1 regions, UK, 1998-2007 and 2010-2016
35. Labour Productivity since 2007
-15%
-10%
-5%
0%
5%
10%
15%
20%
25%
30%
-15% -10% -5% 0% 5% 10% 15% 20% 25% 30%
Real GVA Growth
Growth in Hours Worked
Growth in GVA versus hours worked, NUTS 2, 2007-2016.
Decrease in
Labour
Productivity
Increase in
Labour
Productivity
Chart from ONS Regional and sub-regional productivity in the UK: February 2018
East Anglia
41. Net Fiscal Balance by country and region, 2016/2017
(£million)
• In the financial year ending (FYE)
2017, all countries and regions except
London, the South East and the East
of England had a public sector net
fiscal deficit.
• This was the same when North Sea oil
and gas revenue was allocated to
regions on both geographic and
population shares (the chart shows
the geographic share).
-40,000 -30,000 -20,000 -10,000 0 10,000 20,000 30,000
London
South East
East of England
South West
East Midlands
North East
Yorkshire and the Humber
West Midlands
North West
42. Net Fiscal balance from 1999 to Financial Year ending 2016
(£ million)
-60,000
-30,000
0
30,000
60,000
90,000
120,000
150,000
180,000
UK Greater South East (London + SE +EE)
Rest of England (NE + NW + YATH + EM + WM + SW) N.Ireland/ Scotland/ Wales
43. Exports of Services
0 10,000 20,000 30,000 40,000 50,000 60,000
Construction
Retail (excluding motor trades)
Primary and utilities
Other services
Administrative and support services
Wholesale and motor trades
Manufacturing
Insurance and pension services
Transport
Information and communication
Travel
Real estate, professional, scientific and technical
Financial
Value of Services Exports £(m)
East of England London REST of UK £m
So:
the Office for National Statistics (ONS) which is the executive office of the UK Statistics Authority and the largest producer of official statistics in the UK
the Office for Statistics Regulation (OSR) which is the regulatory arm of the UK Statistics Authority. It assesses official statistics for compliance with the Code of Practice for Official Statistics, reports on system-wide issues and on how statistics are used, celebrating when the standards are upheld and challenging publicly when they are not.
And other departments who produce their own statistics.
Full range of our stats.
Most popular? Baby names.
Now turn to transformation.
Thinking back to that circular flow of GDP, can measure in three ways. In theory they are the same. In practice not quite.
New data sources and technology allow us to measure better the economy in each of the three ways.
Expenditure: sales data
Output: VAT (more later)
Income: PAYE (more later)
Emphasise speed, coverage granularity. Important for e.g. Brexit work, industrial strategy.
Emphasise speed, coverage granularity. Important for e.g. Brexit work, industrial strategy.
Also mention ESAS
Charts shown in the presentation cover the latest available data. Many of the charts are taken from economic commentary and statistical bulletins for each topic which are available on the ONS website.
Search “economic commentary” or topic title.
Time series chart of GDP growth
Blue bars represent quarter on quarter (LHS, %)
Yellow line represents quarter on same quarter a year ago (RHS, %)
UK gross domestic product (GDP) in volume terms was estimated to have increased by 0.6% between Quarter 2 (Apr to June) 2018 and Quarter 3 (July to Sept) 2018.
All four sectors of output contributed positively to growth in Quarter 3 2018, with the largest contribution from the services industries at 0.3 percentage points.
Household spending grew by 0.5% while business investment fell by 1.2% between Quarter 2 and Quarter 3 2018.
Net trade contributed 0.8 percentage points to GDP growth in Quarter 3 2018, driven by a 2.7% rise in exports, while imports were flat.
In this chart –UK is the yellow line.
The path of growth since the recession shows that Italy, Japan and Germany experienced larger peak-to-trough falls in GDP than the UK, Canada, France and the US. Of the countries recovered to their pre-crisis levels of GDP, the UK took the joint-longest along with Japan. Italy has yet to recover to its pre-crisis level. Over 40 quarters (10 years) after the start of the recession, the chart shows the US and Canada have grown the most since the recession, followed by Germany and the UK.
The economic well-being dashboard is published within the Economic Well-being release.
Figures above illustrate economic well-being indicators, UK, Quarter 2 (Apr - Jun 2018). Next release 8 January.
Useful info-graphic style dashboard which shows the trends in each variable, and most also have a green or red arrow showing whether they are moving in a positive or negative direction
The variables on the dashboard generally take a household or per capita perspective, for example, on the top row we have GDP per head, net national disposable income per head. There are also wealth characteristics in the dashboard.
I’d like to draw your attention to the middle chart on the 2nd row - which is a behavioural question. This uses data from the Eurobarometer Consumer Survey, conducted by GFK on behalf of the European Commission and asks about people’s perceptions of their financial situation.
So, again I recommend you have a look at the economic wellbeing release each quarter if you are interested in monitoring these wider measures of economic performance.
Chart shows different measures of prices over time. The yellow line is input PPI, which grew by 10.0% in the 12 months to October 2018, slowing from 10.5% in the 12 months to September 2018. The dark blue line shows output PPI which can be thought of as factory gate prices. Output PPI grew by 3.3% in the 12 months to October 2018, up from 3.1% in the 12 months to September 2018. The largest upward contributions to the increase in the annual rate came from petroleum, and clothing, textiles and leather.
The blue bars show CPIH which is consumer prices with owner occupiers’ housing costs included. CPIH was 2.2% in October 2018, unchanged from September 2018.
The Consumer Prices Index including owner occupiers’ housing costs (CPIH) 12-month inflation rate was 2.2% in October 2018, unchanged from September 2018.
The large downward contributions to the change in the 12-month rate from food and non-alcoholic beverages, clothing and footwear, and some transport elements were offset by upward contributions from rising petrol, diesel and domestic gas prices.
Other smaller upward contributions came from items in the miscellaneous goods and services, recreation and culture, and communication sectors.
This chart looks at the relative house price growth since the economic downturn across various regions, compared with AWE regular pay growth and CPIH.
As you can see the regions all experienced a dip in average house prices between January 2008 and May 2009, after which Cambridge (shown by the dark blue line) in particular began to recover fairly strongly. So by the end of 2009 to early 2010, average house prices in Cambridge had already returned to pre-recession levels, and by mid 2013 prices in these areas had also outpaced CPIH and AWE increases since the recession.
In contrast, average house prices in Great Britain (shown by the yellow line) returned to their pre-recession level in mid 2013.
Estimates from the Labour Force Survey (LFS) show the unemployment rate continues its general decline and the employment rate continues to rise generally.
In the three months to September 2018, the employment rate (for those aged 16 to 64) was 75.5%, slightly higher than 75.0% a year earlier – with 31.2m currently employed in the UK. (LHS)
In contrast 1.4 million people were unemployed with a rate of 4.2%.
Compared with the same quarter a year ago, labour productivity, on an output per hour basis, grew by 1.4% and has been growing for the past seven consecutive quarters.
A 1.4% growth, compared with the same quarter in the previous year, is significantly lower than the long period of average productivity growth prior to the economic downturn, and represents a continuation of the UK's “productivity puzzle”.
Productivity in Quarter 2 (Apr to June) 2018, as measured by output per hour, was 17.6% below its pre-downturn trend – or, equivalently, productivity would have been 21.4% higher had it followed this pre-downturn trend
Increase in demand for regional and local statistics to understand the economy and society at a sub-national level
Release date: 26 April 2018
https://www.ons.gov.uk/economy/nationalaccounts/uksectoraccounts/compendium/economicreview/april2018
Differences in types of industry that have created growth pre-and post recession.
Only half of 40 NUTS 2 regions had increase in real productivity 2007-2016
NI 1% increase in GVA, 2% decline in hours worked.
NOTE APS/LFS show slightly higher hours worked growth than prod hours.
Release date : February 7th 2018
Next release date: TBA
City region, LEP, NUTS2, NUTS3
Real data – NUTS2 only
Release date: 26 April 2018
https://www.ons.gov.uk/economy/nationalaccounts/uksectoraccounts/compendium/economicreview/april2018
This Figure shows firm-level productivity data for the non-financial business economy for regions and countries.
It shows the proportion of firms at different levels of gross value added (GVA) per worker.
The distributions are skewed to the right, indicating that in all the regions there are fewer firms with high productivity levels than firms with lower productivity levels.
London clear firm level productivity advantage in both service sector categories.
Release date: 26 April 2018
https://www.ons.gov.uk/economy/nationalaccounts/uksectoraccounts/compendium/economicreview/april2018
Release date: 26 April 2018
https://www.ons.gov.uk/economy/nationalaccounts/uksectoraccounts/compendium/economicreview/april2018
Release date: 1 August 2018
Next release: TBA
Country and regional public sector finances: Financial year ending March 2016
https://www.ons.gov.uk/economy/governmentpublicsectorandtaxes/publicsectorfinance/articles/countryandregionalpublicsectorfinances/2015to2016
https://www.ons.gov.uk/economy/governmentpublicsectorandtaxes/publicsectorfinance/articles/countryandregionalpublicsectorfinances/2016to2017
Release date : 24 October 2018Next release : TBA
https://www.ons.gov.uk/businessindustryandtrade/internationaltrade/datasets/regionalisedestimatesofukserviceexports
Local Authority GVA for all LA’s in the Cambridgeshire and Peterborough Combined Authority:
LA’s Cambridge, East Cambridgeshire, Fenland, Huntingdonshire and South Cambridgeshire make up NUTS3 region Cambridgeshire CC (East of England)
Peterborough is in the NUTS 3 Peterborough (North East )
Regional Gross Value Added (Balanced) by Local Authority in the UK – Trevor Fenton
Release date : 20th December 2017
Next release : December 2018
Jan-Dec 2015
The spatial distribution of industries in Great Britain: 2015
Regional and local industrial specialisms and the spatial distribution of industries in Great Britain
Release date: 28 March 2017
Location quotient data and industrial specialisation for local authorities
Location quotients can be interpreted as a local measure of geographical concentration of industries. They are calculated as the quotient between the local share of employee jobs in a specific industry and the local share of national employee jobs. LQ = (Ei,r / Ei ) / (Er / E)