Big Data on Vessel Traffic for Nowcasting Trade Flows in Real-Time
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Big Data on Vessel Traffic for
Nowcasting Trade Flows in Real-Time
Serkan Arslanalp Marco Marini
Patrizia Tumbarello
Statistics Department
International Monetary Fund (IMF)
New Techniques and Technologies for Statistics 2019
Brussels — March 12, 2019
The views expressed in this presentation are those of the authors and do not necessarily represent the views of the IMF, its Executive Board, or IMF management.
Outline
Motivation
Key Questions and Findings
Data and Methodology
Validation Study
• Maritime statistics
• Trade statistics
• Direction of trade statistics
Relevance for Policy
2
Motivation: Explore Innovative Sources to Complement Official Statistics
This work is aligned with:
IMF’s Overarching Strategy on Data and Statistics (2018)
IMF’s Staff Discussion Note on Big Data (2017) to provide innovative,
real-time, and granular insights
Benefits of Big Data on Vessel Traffic:
Possible improvements of timeliness and periodicity of official trade data
Provide additional granularity on trade flows
Challenges:
No internationally accepted methodology Validation study using Malta as
benchmark
Access to data Multiple data providers (e.g., MarineTraffic.com)
3
Key Questions and Findings
Can Vessel Traffic Data track Official Maritime Statistics?
Yes. Vessel traffic data appear to show higher coverage of ships than official port statistics
for Malta
Can Vessel Traffic Data be used to Nowcast Official Trade Statistics?
Yes. Good correlation with movements of official trade data for Malta
Can Vessel Traffic Data Offer a Breakdown of Trade Statistics by Partner Country?
Partly. Vessel traffic data are affected by shifts in maritime routes and transit trade
4
Data and Methodology
5Automatic Identification System (AIS) Data for Real-Time Tracking of Ships
Automatic Identification System (AIS)
allows for real-time tracking of
commercial vessels
An “air traffic control system” for ships
Introduced in 2004 after the International
Maritime Organization (IMO) made it a
requirement for major commercial vessels
for safety reasons
Real-time data tracked by several data
providers and made available online
6Visualizing Vessel Traffic Data
◼ For a visualization of the data, play this video
7Port Calls: Fusion of Vessel Positions and Port Boundaries
Defining Port and Anchorage Boundaries:
Port of Cartagena, Spain
AIS messages, when aggregated, contain
billions of ship positions and other voyage-
related information in real-time
Port call data are generated combining
positional data and geofenced areas. Port
calls focus only on vessel activity near a port,
particularly on incoming and outgoing vessels
Defining port and anchorage boundaries
requires careful work. More than 7,000
ports being monitored by MarineTraffic
8Key Contribution of the Paper: Filter to Identify Ships Involved in Trade Activity
Filtering Genuine Cargo Shipments Based on Duration of Stay at Port
(Number of ships by duration of stay in Marsaxlokk and Valletta. Period: 2015-2018)
6000
18.1%
4500 15.2%
Focus on: 13.1%
Number of Ships
11.9%
Container and general cargo ships 3000 9.0% 8.6%
Stays at port between 5-60 hours 7.9%
5.7%
1500
3.5%
2.5%
2.0%
1.4% 1.2%
0
0-4.99
5-9.99
10-14.99
15-19.99
20-24.99
25-29.99
30-34.99
35-39.99
40-44.99
45-49.99
50-54.99
55-59.99
>60
Filtering (exclusion) rules:
Duration of stay at port (hours)
Ships in transit
Anchorage and bunkering tankers Sources: MarineTraffic, Staff estimates.
Missing data pairs (e.g. ships arrived but not departed)
Stays in port that do not reflect trade activity (i.e. drifting or repair/maintenance) 9Three Indicators of Cargo Activity
Cargo number indicator: number of cargo ships (filtered)
Cargo size indicator: sum of gross tonnage of cargo ships (filtered)
Both Indicators are Comparable with
Official Maritime Statistics
10Three Indicators of Cargo Activity (continued)
Cargo load indicator:
is the deadweight tonnage of ship i arrived in port on a given week t
is the reported draught of the ship upon arrival
is the reported draught upon departure
is the maximum draught reported by the ship in the sample
Draught
Proxy Indicator of Trade Volumes in Goods
(Sum of Imports and Exports)
11Validation Study
12Why Malta as Benchmark?
Countries with Largest Shares of Seaborne Imports in the EU
(Extra-EU Imports by Mode of Transport, percent of total extra-EU imports)
Period: 2016.
Denmark 56.1
Small open economy Netherlands 61.0
Slovenia 61.2
Lithuania 61.2
Most international trade is seaborne
Finland 64.5
Italy 65.6
High-quality official statistics available on Malta 69.8
Eurostat portal can serve as benchmark Spain 73.4
Greece 76.0
Portugal 78.7
Sea Air Road Rail
Source: Eurostat.
13Our Cargo Number Indicator Broadly Tracks Official Maritime Statistics,
Highlighting a Larger Coverage of Ships
Malta: Cargo Number Indicator Good Proxy of Official Port
Malta: Cargo Number Indicator (weekly)
Number of Ships arrived using AIS-based port calls. Period: 2015-2018.
Statistics Cargo number indicator vs. Official number of ships arrived in Marsaxlokk
and Valletta. Period: 2015-2018, quarterly.
140 1600
130 1500
120 1400
110 1300
100 1200
90 1100
80 1000
70 900
60 Original 800 Official AIS Port Calls
50 5-term Moving Average 700
40 600
2015-1 2015-27 2016-1 2016-27 2017-1 2017-27 2018-1 2018-27 2015Q1 2015Q3 2016Q1 2016Q3 2017Q1 2017Q3 2018Q1 2018Q3
Sources: MarineTraffic, Staff estimates. Sources: Eurostat, MarineTraffic, Staff estimates.
14High Correlation with Official Trade Volumes
Malta: Cargo Load Indicator (weekly) Malta: Cargo Load Indicator Highly Correlated with Official Trade
Deadweight tonnage adjusted for draught in AIS-based port calls. Index 2016=100. Volume (correlation 0.65)
Period: 2015-2018. Cargo load indicator versus official trade volume index, 2016=100. Period: 2015-18
180 140
160 130
140 120
120 110
100 100
80 90
60 80
40 Original 70 Official AIS
5-term Moving Average
20 60
2015-1 2015-27 2016-1 2016-27 2017-1 2017-27 2018-1 2018-27 Jan-15 Jun-15 Nov-15 Apr-16 Sep-16 Feb-17 Jul-17 Dec-17 May-18 Oct-18
Source: MarineTraffic, Staff estimates. Sources: Eurostat, MarineTraffic, Staff estimates.
15AIS Data Provides Breakdown by Partner Country that is Affected
by Maritime Routes and Transit Trade
Malta: Imports by Country of Origin (Customs Data) Malta: Imports by Country of Last Port (AIS)
Share by Partner Country based on Direction of Trade Statistics. Period: 2015-2018. Share by Partner Country based on Cargo Load Indicator. Period: 2015-2018.
100% 100%
90% 90%
80% 80%
70% 70%
60% 60%
50% 50%
40% 40%
30% 30%
20% 20%
10% 10%
0% 0%
2015 2016 2017 2018 2015 2016 2017 2018
Italy Germany United Kingdom France Spain Rest Italy Egypt Spain France Turkey Rest
Sources: IMF Direction of Trade Statistics. Sources: MarineTraffic, Staff estimates.
16Relevance for Policy
International trade has an important share of GDP in many countries. Tracking a
country’s trade in real-time may offer prompt and informative insights on economic activity
The methodology can be extended to other countries, particularly those with a significant
share of trade carried by ships
On a global scale, vessel traffic data offers great potential for the monitoring of world
trade flows on a real-time basis, which would be an invaluable input for multilateral
surveillance.
Our results do not question the accuracy of official trade statistics but offer ways to
enhance their timeliness, periodicity, and granularity
17Application to Small States: Long Lags in Timeliness/Periodicity of Official Trade Data
Timeliness of Merchandise Trade Data Periodicity of Merchandise Trade Data
(in percent of all small states) (in percent of all small states)
40% 50%
30% 38%
20%
25%
10%
13%
0%
1-2 months 3-5 months 6-11 months 12 or more 0%
months Monthly Quarterly Annual
Source: Authors’ calculations. Source: Authors’ calculations.
Note: Sample includes all small state members of the IMF (43 in total).
Small states have population of less than 1.5 million people.
18Big Data on Vessel Traffic for
Nowcasting Trade Flows in Real-Time
Thank you!
For further information:
Serkan Arslanalp (sarslanalp@imf.org)
Marco Marini (mmarini@imf.org)
Patrizia Tumbarello (ptumbarello@imf.org)You can also read
