# Tag Info

13

Most contemporary NN systems are just made to use the raw price time series for input (maybe with some kind of simple normalization), but for my thesis I wrote a system which traded equities with an ANN with technical indicator inputs (MAs, MACD, even pattern matching for stuff like Head-Shoulders, support levels, etc.). So at least conceptually it's ...

12

Not saying this trade won't work, but there's certainly no guarantee that it will... Given that QE will stop in October is well teleported at this point and has been expected since last year, you'd think this should be fully priced in. Last year, when the "tapering" talk started, Treasuries did sell off quite a bit, but has since rallied all the way back. ...

12

This is a basic fact about futures trading and the storage of commodities. The phrase that was used by futures traders in the old days (and probably still today) was "the contango is limited by the carrying cost, there is no limit to the backwardation". This means that for example if spot gold is at 1200, gold dated one year from now cannot possibly sell ...

12

We assume that, under the probability measure $Q$, \begin{align*} dS_t &= S_t\big(r_t dt + \sigma dW_s(t)\big),\\ dr_t &= -k\, r_t dt + \alpha dW_r(t),\tag{1} \end{align*} where $d\langle W_s(t), W_r(t)\rangle_t = \rho dt$. From $(1)$, for $s\ge t$, \begin{align*} r_s = e^{-k(s-t)}r_t + \alpha\int_t^s e^{-k(s-u)} dW_r(u). \end{align*} Then, for $T\ge ... 11 Using the following data from 12/18/16: Jan 2017 Fed funds futures =9936, Jan 2018 Fed Funds futures =9877 implies that 99.36-98.77 = 59bp of hikes are built in for 2017. IF you assume the only two possibilities are 2 hikes or 3 hikes (meaning, 50bp or 75bp of hikes, assuming each hike would be 25bp), then by simple linear interpolation the probability of 3 ... 10 Treasury futures are actually really complicated... There are complete books dedicated to this topic (e.g., The Treasury Bond Basis) and really good sell-side research papers ("Understanding Treasury Bond Futures" by Salomon Brothers) that I highly recommend. You're actually very much on the right track, but I'll try to paint a somewhat complete picture. ... 10 I will talk about equity futures. Commodity futures can be slightly different, as I briefly point out. Equity futures are standardised exchange-traded instruments. Futures on stock indices are especially liquid. The reason for that, is that one cannot simply buy/sell an index as he would buy/sell a single stock because an index is merely a reference value, ... 9 You are missing the futures basis and roll cost. Futures expire, and need to be rolled into the new expiry. The basis is not static and can vary considerably, depending on the specific underlying and contract. Quants may have a hard time to appreciate this but the basis is not at all fully quantifiable at all times: It can hugely vary entirely due to shifts ... 9 The best theoretical model for pricing vix futures and options is a variance gamma model. However in practice that class of models is difficult to get robust results... In practice, most floor traders in vix products base their hedging off of the SPX option chain. Vix is calculated from those options, in the first place, so this approach makes intuitive ... 8 Not really a quant question, but a quick search led to this from the CME: http://www.cmegroup.com/market-data/files/CME_Group_Settlement_Procedures.pdf. Unfortunately it depends on the contract, for example: Equity Futures: For S&P and NASDAQ, the settlement price of the lead* month contract is the midpoint of the closing range determined based on pit ... 8 VIX is calculated from a basket of SPX options, and VIX futures expire into following expiration, e.g. September VIX futures that will expire next Wednesday will use SPX October options chain to calculate settlement value. If$B$is the value of the basket then VIX value at expiration is$\sqrt{ B }$. Then VIX futures price is the expectation of the basket$...

8

The predictor variables would consist of the input layer to the neural network. The output layer would consist of your target. You need to specify the hidden layer, number of nodes per layer, the learning algorithm, and the learning algorithm stopping criteria. Typically inputs are normalized (first-differenced, z-scored, etc.) before inputting into the ...

8

Just take something like $$\frac{\log{\frac{F_j}{F_i}}}{t_j - t_i} \times 365$$ where $t_i$ denotes the expiry (or alternatively delivery) date of future $i$. The annualization is so you can compare different futures.

8

Forward delta is 1 (defined as change in the value of the forward with respect to an instantaneous change in the price of the underlying, holding everything else constant). However for a meaningful discussion of the differences in forward and futures pricing, the forward price delta of forwards should be considered and it is exp(r(T-t)).Though the delta ...

8

There is actually a lot of art involved. The most simplistic framework is as follows: The first step is to obtain a list of FOMC meeting dates. These are available currently for 2015 and 2016 here. If you're interested for rate expectations beyond 2016, you'd need to "guess" the meeting dates in the future based on past patterns. The next step is to ...

8

Treasury bond futures are surprisingly complicated - this is an attempt at a short explanation, it will obviously gloss over some details, but hopefully gives you a flavour of how they are priced. The most important fact is that the underlying is not a single bond, but a basket of bonds. For example, the US Treasury Bond Futures contract spec says that you ...

7

For reference, note that execution strategies for some types of futures contracts can be very different from equities. An example are Interest Rate Futures, e.g. here. The main reason lies in microstructure differences. For some more details see the white paper "US Treasury Futures Roll Microstructure Basics" by Quantitative Brokers (I have no affiliation ...

7

CXO-Advisory investigate the claim of this paper and conclude that evidence indicates that changes in open interest in futures markets are strong predictors of returns for associated asset classes, even after controlling for a number of conventional predictors. They state that investors may be able to exploit these predictive powers via tactical asset class ...

7

Think of it like a forward trade on the settlement price. If you are buying with a TAS you are agreeing to go long the futures contract at the settlement price (+/- the offset), and whoever you trade with is agreeing to go short at the same price. It is guaranteed because the exchange becomes the counterparty for both traders and there is a margin deposit. ...

7

Your question is an important one, but I am not aware of any particularly satisfying answer. There are several papers on this issue -- see Luo and Zhang 2009 and Zhang et al 2010, just for example. One thing to note is that VIX futures are not always in contango -- after large jumps in the VIX, they can even be in rather steep backwardation. I have heard ...

7

Your questions about contango in VIX futures have close analogies in options too. The Black & Scholes model suggests that all time frames and all strikes should have the same implied volatility, but they don't. I think one of the reasons is that the B&S model assumes that stock returns are distributed in a normal (gaussian) distribution, but ...

7

VIX is a measure of volatility -- something that changes explicitly with uncertainty. The chances of uncertainty arising tomorrow, is lower than the chances of uncertainty increasing in the longer term. A long-dated option should therefore have more "potential uncertainty" baked into the price. When pricing normal futures, the price is a martingale, the ...

7

This part of your post In addition, on expiry day the holder (...) is wrong. [Short Story] Due to the daily variation margins calculated by the clearing house on each market close, you have already received/coughed up what you should upon expiry. If the contract is cash-settled, the story thus ends here. In case of physical delivery however, although ...

7

Personally I hate the term "roll cost" and prefer "roll yield" or "effect of rolling". It is not really an out of pocket cost (it involves no outlay or receipt of cash). It has to do with contango and backwardation. When you close the contract that expires soon, it is priced close to Spot, but the new contract that you enter into may be priced above or ...

7

Based on the your comments, I believe the issue lies with what you consider to be "carry." The reality is that there's no consensus. So let's take mini steps. We'll start with what rates guys consider as "pure carry." In this most classical and fairly strict definition, carry is the deterministic component of expected returns – you know exactly what it is ...

7

This is a surprisingly complicated question that encompasses many moving parts. Without knowing exactly what your objectives are, it's a bit difficult to offer concrete advice, so I'll provide some general comments below. Mechanically, you earn the total return when you buy and hold a real bond or a bond ETF. By contrast, bond futures are financed ...

6

Farmers (usually referred to as hedgers) typically buy Futures, not options. The CBOE does not transact in agricultural commodities The CME, and other exchanges, transact in agricultural commodities. These exchanges grew, for the most part, directly from the trading interactions of farmers(hedgers/producers) and middle-men(speculators), in the area where ...

6

Maybe a better question title is "Can futures market open interest predict commodity, treasury, and equity returns"? I saw this paper in an earlier form and it still baffles me. Superficially, it makes sense that price*quantity holds more information than just price when quantity can change quickly (i.e., outstanding futures contracts changes more quickly ...

6

Short Answer : Futures don't have Greeks Long Answer : Assuming a non strictly mathematical (i.e. false) point of view. Well, having Greeks on VIX Futures is not relevant, VIX value is itself a "Greek" (and Futures don't have Greeks). Sensitivity to Price of the Underlying : Insensitive (ν = 0) Volatility of the Underlying : Delta Δ = 1 (to Volatility of ...

6

As you pointed out there are many ways to adjust for the roll overs. Hence, I guess you would agree that there is no one-size-fits-all answer to this. It really depends on the usage of the data: First think about how the trades in your back test are structured. If they are longer-term trades and you hold over roll overs then think what you would do if you ...

Only top voted, non community-wiki answers of a minimum length are eligible