Transcript to read more

Model to predict oil prices
Address : 477,Udyog Vihar Phase III
Gurgaon (Haryana) -122001, India
Tel : +91 124 4771800
Website : www.i3c.in, Email : [email protected]
This document is solely for the use of client personnel. No part of it may be circulated, quoted, or reproduced for distribution
outside the client organization without prior written approval from i3 Consulting.
Please note that this is a personal opinion and is not a recommendation to do any trading whatsoever based on this report.
Any entity planning to take exposure should do their due diligence before taking any such action
Executive summary
• Oil price has a strong relationship with factors like world GDP growth, speculative
trading, momentum indicators, dollar index and oil price as lead indicator of itself
• As against general perception, following factors do not impact the oil price or do not
have an economically sustainable valid relationship for predicting future price of oil –
(1) Demand supply mismatch, (2) Price for long term futures contract of oil (3) Oil
Inventory levels of OECD and US
• Oil prices could well be modeled precisely using the above stated factors, however the
market sometime deviates from its economic value due to higher speculative activity
which is primarily driven by either driven by positions taken by financial players or
owing to uncertain political environment, esp. US and the Middle East
• Our proprietary model predicts oil to be weaker in the next couple of quarters owing to
negative outlook in world GDP growth, no major reason for higher speculative trading,
and weak momentum indicators, however the market will turn itself – our models
predict price to be in the range of USD 52-USD 57 by end of 2009 and in the range of
USD 68 – USD 75 by end of 2010*
* Please see assumptions before reading about price
GDP growth rate does impact oil price movements
GDP growth with change in oil price
y = 22.387x - 0.2313
R2 = 0.3233
60%
40%
20%
-0.5%
0%
0.0%
-20%
-40%
-60%
-80%
0.5%
1.0%
1.5%
2.0%
Demand supply mis-match does not result in long
term shifts of oil price movements
Portion of “long” speculative trading as percent of
all “long” trading related to oil price movement
2
y = -2666.8x + 1436.5x - 110.18
160
Portion of “short” speculative trading as percent of
all “short trading” related to oil price movement
2
y = -1498.2x + 696.43x - 3.3974
160
2
R = 0.0676
2
R = 0.2107
140
140
120
120
100
100
80
80
60
60
40
40
20
20
0
0
0%
10%
20%
30%
0%
10%
20%
30%
Despite huge volatility in crude price…
Monthly percent change in oil price
Co-efficient of variation is 31% suggesting 3 sigma confidence levels to be 200% of mean value
20%
10%
0%
1-Aug- 17-Feb- 5-Sep- 24-Mar- 10-Oct- 28-Apr- 14-Nov- 1-Jun- 18-Dec- 6-Jul-09 22-Jan04
05
05
06
06
07
07
08
08
10
-10%
-20%
-30%
-40%
Only 30% of
monthly price
variation is
within +- 5%
…Crude price are auto correlated, i.e. historical
prices gives indication of future price movement
Oil price as lead indicator of itself (4 months lead indicator)
160
y = 0.6107x + 29.227
R2 = 0.3888
140
120
100
80
60
40
20
0
0
20
40
60
80
100
120
140
160
The momentum indicators of trading are also
predictive of price movement
6 months moving average related to spot oil price
y = 0.6617x + 25.375
140
R2 = 0.5925
120
100
80
60
40
20
0
0
20
40
60
80
100
120
140
160
We have created a transformed variable to
capture the momentum and the directionality of
bullishness in the market
Indexed momentum capturing its strength / weakness correlated to oil price movements
4
3
2
1
0
0
50
-1
-2
-3
-4
Higher clustering of price
bands based on weaker /
stronger momentum index
100
150
Dollar index has a strong correlation with crude
price but economically its relationship seems to be
unstable
Dollar index correlated to oil price movement
y = -0.1575x + 105.22
R2 = 0.6866
102
100
98
96
94
92
90
88
86
84
82
0
50
100
150
Despite the strong
negative correlation of
oil price with dollar
index, it is a short
term economic
phenomenon which
might undergo slight
change once world
economy and US
economy starts
improving giving
different dynamics to
global demand and
dollar strength
Inventory levels of a country are economically
dependent decisions based on oil price, however can
give direction of short term oil price movements
Oil inventory levels of OECD
Oil inventory levels of US
y = -0.5087x + 2669
R2 = 0.0525
2800
y = -0.7073x + 1068.3
R2 = 0.244
1120
1100
2750
1080
1060
2700
1040
2650
1020
1000
2600
980
2550
960
940
2500
0
50
100
150
0
50
100
150
Futures contract barring 1 month contracts of oil do
not provide forward looking guidance on price
movements
1 month future contract price against spot
Methodology
y = 0.4002x + 50.29
R2 = 0.3314
150
100
50
•
•
0
0
50
100
150
6 month future contract price against spot
Objective of the exercise is to assess
the ability of the futures market to
estimate future spot prices effectively
We compared the future prices
applicable for a period with the spot
price of that period for e.g. 6 months
futures contract prices of Jan 2006
with spot price of August 2006 and so
on
y = -0.1846x + 96.601
R2 = 0.0563
150
100
Insight
50
•
0
0
50
100
150
18 month future contract price against spot
y = -0.1047x + 86.692
R2 = 0.0824
150
100
50
0
0
50
100
150
•
One month futures market mimics near
term future closely however the long
term futures contract i.e. 6 months or
18 months futures miss the spot price
by a big margin
Futures market though brings in some
semblance of sanity during extreme
speculative environment and can be
relied on to assess the directionality in
the short run
Demand supply mis-match or its expectations in
short runs are not huge and have very little
relationship with oil price movement
Demand supply mismatch related to oil price
y = -2.2912x + 73.609
R2 = 0.0188
160
-4
-2
3 months lead expectation of mismatch related to oil
price movement
140
140
120
120
100
100
80
80
60
60
40
40
20
20
0
0
0
2
4
y = -1.7445x + 75.066
R2 = 0.012
160
-4
-2
0
2
4
Relevant variables are combined together to
predict the future oil price movement
Dollar index
Lead oil price
Momentum indicator
• Dollar index is negatively correlated with oil price movements. It
makes business sense as any strenghtening of dollar should
result in lower oil price, however its link in context of economic
growth of EU/Asia vis-à-vis US can slightly impact the
relationship
• Oil price as lead indicator is helpful in predicting its future price
and is positively related to its future prices
• Indexed momentum indicator is positively related to the oil price
movement
Speculative
interest
• Share of speculative interest in trading is positively correlated
to the oil price movement
World GDP growth
• World GDP growth is positively correlated to the oil price
movement
Future prediction of oil price is dependent on
following outlook
Dollar index
Momentum indicator
Speculative
interest
World GDP growth
• Dollar index is expected to move at hover at around 95/96
levels in the short run. It may move down to around 92 by Q1
2010 depending on improved performance by European
economies
• Calculations reflects a weaker momentum right now which
could dampen price in the short run and will be positive as we
approach Q4 2009
• There is no expectation of higher speculation in this time,
however it is expected to start building up towards the end of
Q4 2009 as oil demand moves up in winters
• World GDP will grow by an annualized rate of -1.3% in 2009
and 2.0% in 2010 respectively (this is a combination of
forecasts from IMF and EIU)