I feed www.lmt-expo.com with data from bloomberg. There are quite a few excel add-ins; www.mbrm.com tech hackers, etc...
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Option replication and exotics journal
- Thread starter riskarb
- Start date
- Status
- Not open for further replies.
YW, Rudi.
Regarding the open Nikkei barrier -- I am not throwing any cash at the Nikkei with the yen as strong as it has been. I've bought 10MM $/yen as somewhat of a replication hedge. The rationale being that we have no inversion risk on JPY, but we have considerable risk on delta-inversion in the barrier. Plus, we get paid to hold the fx hedge.
Long 40 SGXNK//long 10MM USDJPY from 114.55 average
Regarding the open Nikkei barrier -- I am not throwing any cash at the Nikkei with the yen as strong as it has been. I've bought 10MM $/yen as somewhat of a replication hedge. The rationale being that we have no inversion risk on JPY, but we have considerable risk on delta-inversion in the barrier. Plus, we get paid to hold the fx hedge.
Long 40 SGXNK//long 10MM USDJPY from 114.55 average
Riskarb, I was hoping to run a hypothetical by you in a fledgling attempt to transfer the hedging idea to the es.
The initial position is 20 Feb 1255x1235p x 1335x1350c iron condors filled at 4 each for a total credit of $20k (commissions excluded). The call spread is a smaller to make the smiles more symmetrical. The exit for the ic is to go flat if either risk strike is hit which would lose an estimated $20k.
Assuming the es is at 1295 at initiation a full 10 cars would be needed to fully hedge the loss if the trade was shut down at 1335 and $20k is lost at this exit point. This is a rough estimate and does not account for discontinuity or theta reducing exit risk.
On the long side, to hedge the trade we would:
Long 1 es @ 1302
Long 2 es @ 1309
Long 4 es @ 1316
Long 8 es @ 1323
Long 16 es @ 1330
At 1335 Long contracts would equal 31 cars for a total of $16,900 and the hedge would be soft $3100 of the estimated $20k loss if an exit was required fairly soon after trade initiation. Theoretically, every day of theta lowers the estimated exit price for this trade.
As far as the exits to avert whipsaws I understand it may need to be an act of discretion. However, using the same intervals as stops to reduce positions may work from a rule perspective. Thoughts?
You seem to be successfully implementing your hedging idea with the exotics. Could this be extrapolated to the es and vanillas in this manner, your preference for flys aside?
Do you have other suggestions?
The initial position is 20 Feb 1255x1235p x 1335x1350c iron condors filled at 4 each for a total credit of $20k (commissions excluded). The call spread is a smaller to make the smiles more symmetrical. The exit for the ic is to go flat if either risk strike is hit which would lose an estimated $20k.
Assuming the es is at 1295 at initiation a full 10 cars would be needed to fully hedge the loss if the trade was shut down at 1335 and $20k is lost at this exit point. This is a rough estimate and does not account for discontinuity or theta reducing exit risk.
On the long side, to hedge the trade we would:
Long 1 es @ 1302
Long 2 es @ 1309
Long 4 es @ 1316
Long 8 es @ 1323
Long 16 es @ 1330
At 1335 Long contracts would equal 31 cars for a total of $16,900 and the hedge would be soft $3100 of the estimated $20k loss if an exit was required fairly soon after trade initiation. Theoretically, every day of theta lowers the estimated exit price for this trade.
As far as the exits to avert whipsaws I understand it may need to be an act of discretion. However, using the same intervals as stops to reduce positions may work from a rule perspective. Thoughts?
You seem to be successfully implementing your hedging idea with the exotics. Could this be extrapolated to the es and vanillas in this manner, your preference for flys aside?
Do you have other suggestions?
That looks sound, and it's certainly applicable to run a martingale into vanilla positions. I closely monitor my hedge b/e if all hedge levels are executed and begin to liquidate the hedge when it reaches b/e on a retracement. Gamma bleed is intense near expiration on the <20d options, so a reversion to b/e on hedge should result in a decent paper-gain on your primary.
The double barrier exotic and the vanilla IC carry similar gamma curvature, so modifying the martingale for analog conditions[risk beyond strike] and gamma mag. is a fairly simple process.
Mimic the gamma slope as closely as closely as possible by doing a "what if" portfolio gamma calc at each of the hedge levels.
The double barrier exotic and the vanilla IC carry similar gamma curvature, so modifying the martingale for analog conditions[risk beyond strike] and gamma mag. is a fairly simple process.
Mimic the gamma slope as closely as closely as possible by doing a "what if" portfolio gamma calc at each of the hedge levels.
Thank you for your feedback, very interesting indeed. I've often wondered if an idea like this would work for a retail trader with a small account but thought it might be cost prohibitive. I'll begin to evaluate it more.
I will say I think having only 7 day trades on would be an advantage as opposed to say a full 4-6 week options cycle on. The last 1.5 weeks in the es is testament to that with the massive whipsaw.
Good job on your nikkei trade.
I will say I think having only 7 day trades on would be an advantage as opposed to say a full 4-6 week options cycle on. The last 1.5 weeks in the es is testament to that with the massive whipsaw.
Good job on your nikkei trade.
Thanks. This Nikkei trade will produce a double on the sub-account since inception of the journal. I will have a PDF of fund performance stats in mid-Feb, which will not include any gains or losses from 2006. Jan gains will be 1.5MM if this payout is received.
Down approx $60k on the Nikkei futures hedge and spot USD/JPY long.
Down approx $60k on the Nikkei futures hedge and spot USD/JPY long.
- Status
- Not open for further replies.