Merry Christmas to all fellow ACD'ers. (Real traders and ole homegamers like me. ) Special thanks to you Mav for sharing the wealth over many years. We all appreciate it very much.
Thanks guys. Merry Christmas and Happy Hanukkah to all. It's great to be a part of one of the longest running and highest quality trading threads on ET.
Regarding storage: I'm attaching a comprehensive white paper (not mine) on this topic. A man by the name of Working (1934) first assembled the theory of storage asserting that a commodity holder has risk in not selling his asset today and holding it for future sale. He is compelled to hold this asset only if he can be compensated for the cost of storing that asset plus interest costs through opportunity cost of capital. This explains the forward curve and why in a "normal" market, the forward curve is equal to the cost of capital plus storage cost. It was found though that the forward market often traded much higher then this forward equilibrium price. It would eventually be given a name called the "convenience yield". This yield was essentially a dividend given to physical asset holders when storage was scarce. If storage was full and anyone looking for storage would be willing to pay a "premium" above the going market rate. This is because if they can't store let's say their oil, they will be forced to firesale their product on the open market at deeply discounted prices. So those that already secured storage earned this additional premium. But in order to capture this premium they had to sell forward at the higher then fair value price and lock it in. By locking it in they did two things: sold their inventory forward at fixed price plus earned the added dividend from the "convenience yield". This convenience yield made sense in contango markets but it did not explain backwardated markets. It turns out that it does. The convenience yield simply inverts from a cash flow to a cost or the sign flips from negative (meaning incoming cash flow) to positive. Now the convenience yield is a cost to the asset holder and the asset takes on the properties of an option. Meaning it has an asymmetric payoff. The "cost" comes from what the asset holder is giving up if he/she chooses to NOT sell their asset in a time of shortage where front prices are selling at steep premiums to further out prices. Since this market is in short supply, buyers will pay anything for it. But there is a risk, just like owning a call, higher prices will eventually bring supply back to the market and near term prices will return to equilibrium. But in the very short term, holding out a little longer can bring much higher profits. So the cost of this "call" option is the foregone profits in return for higher prices. The convenience yield reflects this cost in the backwardated prices. The convenience yield therefore can be modeled like an option. It has a mean reverting stochastic process whose value is ultimately determined by the volatility of the underlying asset. This whole thing is held together by the process of physical arbitrage. If physical asset holders cannot execute the arbitrage, then prices cannot converge. For example, there could be bottle necks in pipelines leading into Cushing or the Gulf coast that prohibt price convergence. Ultimately the forces of supply and demand will respond to any excess risk premiums the market is offering since these premiums are essentially free money to asset holders. This means a large majority of the time, the forward curve will trade in contango at the cost of capital plus storage. The market can trade at a premium in contango without convergence for long periods of time if storage conditions stay full. This acts as an incentive to producers to stop producing and reduce supply until the market re-balances. When markets become backwardated though there is no limit as to high they can go in these conditions. Under scarcity conditions, the market will have to pay whatever the cost to secure supply and pass those costs on to the end user until the end user reduces demand. There is a food chain here but its very slow to act. There is no equivalent arbitrage for backwardated markets. The reason is because in a contango market, inventory holders can shift their asset forward in time from the present to the future via storage. One cannot however bring inventories from the future to the present. Therefore prices have to absorb the burden. It's VERY important to note that these structural opportunities are NOT available to the paper trader or futures trader. They are available only to the physical holder or in the case of storage, to the owner of storage capacity. The futures trader is exposed to a price they cannot hedge. So while you can trade the forward curve via futures and look for mispricings, the physical players can squeeze the paper traders. Remember, the relationship only holds if you can and intend to secure storage capacity. Without this, only you have is part of the leg. You have for example oil that will either be delivered to you at Cushing or you will make delivery to oil at Cushing. You are exposed to variable transportation and storage costs which of course is priced into the very thing you are trying to trade. From this, we can deduce that the forward curve can be modeled by the components that affect it's price which is current storage levels, the volatility of the underlying market and the basis prices which are the cost of moving the product to the next available arbitrage. The attached pdf will provide the math from the above commentary. Hope this answers your question and Merry Christmas!
Hey Mav Really enjoying your thread from over the years. I'm an intraday Eurex trader got interested in ACD a few months ago and read Fisher's book a few times over now. I'm employing ACD with pivots and vwap standard deviation bands. ACD for context, usually vwap bands for entries and ADR high/low and weekly pivots for some of the longer hold exits. I'm using Kase bars to analyse monthly, weekly and daily trends and using monthly and weekly pivots to stay the right side of the market after doing my prep. I see from the thread that you're using monthly and weekly ORs. That seems a very nice way to create an even clearer long term contextual picture. I'm trading the bund and eurostoxx and I was unsure for as an intraday only wanting to use weekly/monthly ORs for context, what time frame the ORs should be around. For Eurex products the cash open will be more significant than the Eurex open, which is one hour later at 08:00GMT so I'll start weekly/monthly ORs from that point. Could you kindly guide me on a solid enough OR time frame for the weekly and monthly? Unsure on A and C values but will do some digging into which formula to use based on my timeframe. I like the idea too of creating a long term OR after a big ECB/Fed announcement whether the fundamental picture changes it - almost like a rest button for those longer term ORs. Thanks again for all your insight on this thread
As a second note I'm using daily ATR(10) x 10% for A and 15% for C values for my intraday trades. It seems to me that 20% or 25% is too wide which would prevent me getting on board any of the decent OR top (after A-up) or bottom (after a-down) touch trades that have been profitable for me in the past. However If I start to get hung on early A trades I may increase that percentage a bit. Journalling will show that up soon. Anyone use similar A and C values? Lastly if anyone is using NT could you kindly recommend a good range indicator to measure weekly and monthly ORs as I can only find daily OR indicators on the net. I do post a fair bit in BMT but will be posting here a few of my trades as I've been very impressed with the level of knowledgeable participants on the thread! Happy trading for 2017!
Happy New Year Mav and all the great folks on here. Wishing you a bumper 2017 and every happiness in life!