Transcript Uses/Misuses of Marine Transportation Statistics

International Maritime Statistics Forum
Annual Conference
Gdansk, Poland
April 2008
Uses/Misuses of Marine Transportation Statistics
Introduction (Types of Data)
Elastic Expectations (Forecasting)
Market by Inference (Forecasting)
Supply/Demand Matching
Hybrid Approach
Types of Data
Primary Data:
Focused actionable data collected for a
specific purpose, usually collected through
focus groups, customer surveys/feedback
surveys…...
Secondary Data: Data that was originally collected for
another purpose, usually collected on
forms, such as vessel manifests, customs
documents, Securities and Exchange
filings, vessel registrations.
Analytical Sins – Secondary Data (pro-investment bias)
Elastic Expectations
Up-turns will continue; down-turns
are temporary!
Market by Inference
Project will succeed because
things (GDP, Trade) are growing,
but even if things aren’t growing,
the project will steal market
from others? Project is good
no matter what happens!
Supply/Demand Matching Trade (demand) is growing so we
need more transportation assets
(supply). Or, transportation assets
are wearing out. In either
case, more assets are required!
Elastic Expectations/Market by Inference
and Forecasting
In the 1970’s, Data Resources Inc. provided analysts with tools
for do-it-yourself forecasts.
U.S. Domestic Trade = 453.455 + 0.119 x (real GDP)
(16.865)
(13.653)
Interval: 1950-1975
R2 = 0.886
Elasticity (% change trade/% change trade) = 0.45
The Forecast
Million Tons
U.S. Domestic Trade,
Actual and Forecast
2000
1500
1000
500
0
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Year
Domestic Trade, Actual
Domestic Trade, Forecast
The Result
Billion
Dollars
40
Water Transportation Vessel Assets,
1960-2006
30
20
10
0
1960
1970
1980
Year
1990
2000
Typical Forecasts
Fang & Cone
Million Tons
U.S. Coastwise Trade,
Actual and Fang-cast
Million Tons
450
500
400
450
350
400
300
350
U.S. Coastwise Trade,
Actual and Cone-cast
300
250
250
200
200
150
150
100
100
50
50
0
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Year
Coastwise Trade, Actual
Tooth
Coastwise Trade, Forecast
0
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Year
Domestic Trade, Actual
Domestic Trade, Forecast
Upper Limit
Lower Limit
Conclusion, Hybrid Approach
Investment Decision
Long Term Customer Commitments
Primary Data Collection and Analysis
Secondary Data Analysis
Bottom Line
Start with secondary data, finish with primary data.
QUESTIONS?
Contact Me
Russell Byington
Chief Economist
Maritime Administration
U.S. Department of Transportation
202 366 2278
[email protected]
Web Site: www.marad.dot.gov/marad_statistics