This is an overview of these micro spreads I talked about in a different thread.Thanks RealMoney, you're right to hedge properly really need to approach it from risk perspective. Cointegration just looks at the stationarity of the time series. This thread has saved me time writing a bunch of code to automate it.
I'll put my time and effort into automating a directional strategy, maybe take a look LSTM's for a classification algo, or some of the new Time Series ML models. Test, test, test.
Thanks again!
__ Buy MES / Sell MYM ~ Synthetic Nasdaq 100 (in a 1:1 ratio)
Buy MNQ / Sell MES ~ Synthetic Tech Index (1:1)
Buy MES / Sell MNQ ~ Synthetic S&P 500 (2:1)
Buy MNQ / Sell MYM ~ Growth/Value Spread (2:3)
Buy MYM / Sell M2K ~ Market Cap Spread (2:3)
Buy MES / Sell M225 ~ S&P / Nikkei index spread (1:1)
Thanks again!From what I understand about native spreads listed on CME, the problem is of priority in the matching engine. Imagine if there was a native index spread market on GLOBEX. That would mean (theorizing) that FCMs would have the ability to post/cancel limit orders (in a queue) by specifying a price at which they are willing to clear. Additionally, FCMs would be able to execute market orders on the native spread book. This would create several issues. The most important of which is that the exchange would have to give priority matching to a native spread book.As far as I know, none of the spreads listed above by Real Money is exchange traded. CME provides margin credit for 1:1 ES/NQ but we have to leg into it. Please correct me if this is not the case. I have not been able find the exchange traded pair for ES/NQ in any ratio. Thank you!
"Exchanges don't generally offer native spreads on index futures, not even in the calendars AFAIK." so does that mean legging risk? don't think so have not heard anybody talking about it ( only legging risk that is discussed online is Autospreader doing synthetic spreads. arb spreads )From what I understand about native spreads listed on CME, the problem is of priority in the matching engine. Imagine if there was a native index spread market on GLOBEX. That would mean (theorizing) that FCMs would have the ability to post/cancel limit orders (in a queue) by specifying a price at which they are willing to clear. Additionally, FCMs would be able to execute market orders on the native spread book. This would create several issues. The most important of which is that the exchange would have to give priority matching to a native spread book.
Currently, the best bid/offer on GLOBEX for a particular index future and settlement date is unrelated to any other futures market, and for good reason. Exchanges don't generally offer native spreads on index futures, not even in the calendars AFAIK.
Institutions that need to trade size in an index spread market can get it in the swaps market OTC, or they could use a combination of options, swaps, forwards, futures, etc.
Major players likely wouldn't be using SPAN margin for a trade like this, nor would they want to.
They will have custom risk margining at an investment bank where they can use privately negotiated collateral and funding agreements, cross margining etc. This would enable them to hold the physical asset against an equity swap, for example, affording them more (and more flexible) leverage.
Ok but not even Calendar spreads as Native? can a FCM rep clarify this pleaseThere is NO native/exchange (CME) listed spreads in index futures (ES, NQ, RTY, YM). Many reasons are explained by Real Money in the previous post. Since there is no native spread with its own symbol, there is legging risk. Legging risk can be a little reduced with Autospreader like order handling software. Even Autospreader has features like hung legs, incomplete legs, split orders, etc to deal with legging risks.
Hang on though but is there a legging risk at all specially for Calendar spread of same futures underlying?About this "legging risk": why not just enter with a limit on the costly to fill leg, then market in on the other? You will not get the exact price, but sometimes it will go in your favor, sometimes against you. It's a wash in the end, unless you are a HFT/scalper of course.