• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

Starting Crude Oil and Coal Technical Analysis

To continue my topic,

Here are the following steps I plan to take in the next few days to better grasp the mechanics of Oil pricing:

1) Take daily values of WTI, Brent from bloomberg (i dont have an application for this so i'll use their website)
2) Make a mysql database where I could store it (besides excel worksheet)
3) read on more books on modeling/TA
4) try to fit some curves and use R project's forecast package.

Does that sound okay?

You want to forecast prices? Just start of by understanding inter market effects on oil. Also, remember how the delivery works for oil. It's January...but everyone is trading the March futures because february was already in delivery. You usually trade the month that is two months ahead due to delivery.

Grab some 20 day 15 minute data and see how the market moves intra-day. Look at some daily bars for a year and see how that moves.
 
are you sure of what you are talking about?

Also, remember how the delivery works for oil. It's January...but everyone is trading the March futures because february was already in delivery. You usually trade the month that is two months ahead due to delivery.
 
the two months ahead part of what you said

Ah. That's just the delivery period before expiry. Normally you can trade futures till expiry but with a lot of the commodities you cannot due to delivery and have to roll over or close before delivery period starts. I was just giving the example of right now basically. It's january and we are trading March instead of Feb. Just an odd fact.
 
Alright.

Newb question here.

Is it okay for me to just check the futures prices of oil and say that,
"ahh the spot prices will follow this pattern in the near future"
 
Alright.

Newb question here.

Is it okay for me to just check the futures prices of oil and say that,
"ahh the spot prices will follow this pattern in the near future"

I Lol'ed
 
Dragunov, as I understand it, you work for a generator. Usually, generators sell parts of their available energy into the grid at market, but also have wholesale agreements with retailers and large customers in place. These are usually swap-type structures where the generator would receive fixed $/ per MWh and dispatches fixed volume as per the agreement. In addition you'd also have some purchase agreements between the entity operating the plant and those financing it.

Your risk factors are thus energy, coal, gas, and oil prices plus any systemic ones such as outages due to grid or generation issues. It becomes interesting when you have storage facilities for your fuel sources at your disposal allowing you to do carry trades on the risk factors' forward curves and spot. Often municipalities would lease out empty, underground salt fields connected to different pipelines where you could pump gas in and out. There are a few IBs that specialise on leasing those facilities from the municipalities to trade the gas forward curve (dunno about oil).

At any rate - if you havent done this already - before delving into trading your cost risk factors (ie fuel sources), you'd be better off building a model of your generators to get an understanding of their sensitivity with respect to these and other parameters (like minimum stable generation, ramp rates, heat rates, start up cost dependent on heat state and so forth). Also, you can do pseudo optimisations in regards to your optimal dispatch schedule given market prices and forward curves. The latter being your initial condition (eg Hull and White model for interest rates comes to mind). With a good and well calibrated model, you'd be surprised as to how well the model can reflect reality.

Once you've done the above, I think it'd be much easier for you to think about how to go about trading. Personally, I havent heard of generators which trade the fuel risk factors as a source for profit, but rather to hedge the output and profitability of their plant.
 
Hi Tobias :)
These are usually swap-type structures where the generator would receive fixed $/ per MWh and dispatches fixed volume as per the agreement.

There are, we call them bilateral contracts in our market. (Philippine Luzon grid)

you'd be better off building a model of your generators to get an understanding of their sensitivity with respect to these and other parameters (like minimum stable generation, ramp rates, heat rates, start up cost dependent on heat state and so forth).

True. I use two tools for this. One of them was designed by IAEA, it's called the W.A.S.P.-IV (proprietary optimization software for power grid expansion and costing)


Once you've done the above, I think it'd be much easier for you to think about how to go about trading. Personally, I havent heard of generators which trade the fuel risk factors as a source for profit, but rather to hedge the output and profitability of their plant.

I think i'm at that point already. My goal is to learn how to observe the market. My market buys oil from Singapore indexed through the MOPS. However, I'm only capable of storing the daily prices, without knowing much about what to do with them.

If my boss asks me, "Hey should we fill our storage tanks now, based on your price projections, in the next two months or not?", I can't answer it properly, all I have are data on prices, with no analysis done on them.

I think I'm confusing everyone in this thread, i'm sorry about that. I'm not "trading oil" i'm just trying to gain more knowledge on how to read the price movements.
 
If one aspect of your problem is the valuation of fuel storage and particularly the timing of it, look into the gas storage (real option) and swing contracts modelling approach.

Your storage tanks will have their own physical restrictions like injection and withdrawal rates (usually dependent on level of fuel in the tank), overall capacity,... and market constraints such as lack of liquidity in particular at the back end of the curve...

There are a number of ways how these can be modelled in terms of actual trading strategies such as fully intrinsic (using the initial, observed forward curve only - very straightforward), rolling intrinsic (more advanced but still fairly straightforward if you limit yourself to historical backtesting only) and extrinsic valuation. The latter is usually done using Least Squares Monte Carlo or Tree based methods (though your tree evolves into a forest really when you need to condition on the fuel level state space as well).

This is a very good paper going into extrinsic valuation:
http://web.mit.edu/jaillet/www/general/swing-last.pdf

It's actually on swing contracts, but very similar principles apply to modelling fuel storage.

Taking into account the physical constraints of your tanks, I'd say the intrinsic method is a good, back-of-the-envelope type linear optimisation that'd give you the most optimal injection and withdrawal schedule based on your observable forward curve and represents the lower bound on the value of your facilities.
 
I'm sorry guys, but the topic is gonna have to take a backseat for now.

I'm tied up till feb. 11 (last day of questions submission for GARP ERP exams).

o_o so hard to think of questions...
 
Back
Top