TMC #7! Markets are nothing else than a mechanism to aggregate a bunch of probabilistic scenarios and weigh them into one price: so, what are we pricing in today?
Hi Alfonso, your website is teaching me a lot- thank you! Does the pricing for long term UST yields basically tend to follow the cumulative Forward OIS rates for the corresponding time period e.g. UST 10y is basically al the forward OIS rates from 0 to 10 years? Thanks!
Yes, plus the term premium which is a theoretical premium an investor would require to hold long-term bonds rather than buying and rolling short-end bonds over and over.
It would be great if you could share how you calculate the implied probabilities in more detail as suggested in other comments below. Keep up the good work! Really enjoy your substack
Hi Alfonso, very helpful post indeed. Are you using 1M forward rates here such as USD OIS Fwd Swp 6Y1M etc? Such as S0042FS 6Y1M BLC Curncy ticker in Bloomberg would allow to see history of it in addition to ICVS? Thanks
Enjoyed reading all your articles Alonso! Given that the option-implied probabilities were backed out based on the cost of various spreads, would the sum of the option-implied probabilities at all strikes add up to 1?
Thank you for the great post. Wondering if you could share the method used in calculating the implied probability. It would be great if you could shed some light on whether such method can be used effectively in the FX market.
Hi Hafiz. I gave some hints to Paul, see my comments to his reply. I'm thinking about writing a more technical post in the future describing the process in details. Yes, I am pretty sure you can do the same with the FX market as well.
Wonderful article Alonso. One question and one conment.
You use the option market for equities but not for rates ( where you use OIS). Why?
The problem with using options for equities is that will introduce bias in your estimates. A delta one long position hedged with puts is tax advantaged relative to a sale transaction that generates ( taxable) capital gains.
For rates, I am using OIS forward rates because they are pretty liquid also for maturities down the road. I am not sure I can say the same for Fed Funds or Eurodollar future options - sometimes you don't even have the underlying Eurodollar future contract so many years down the road.
You could use short dated options on deferred contracts. They are very liquid.
For the equity market I use a projection of ERP ( a 3 stage dividend discount model based Equity Risk Premium). That would suggest there is at least another 5-10% to go even if you allow long rates to rise 50 bps from here.
Absolute gold, thanks Alfonso. You said you wouldn't bore us, but actually I'm really interested in calculating option-implied probabilities like you did to produce your chart. Please could you provide a pointer to get started? Thanks again!
You can do this in several ways: one idea would be to use call spreads or digital options for different strikes and see how much you have to pay for a payoff within a certain boundary. That gives you an idea of how much probability the option market is assigning to a certain outcome within a range of strikes.
There are other ways as well: maybe I should write a more technical post about this once.
first of all, it is a real pleasure to read your posts: awesome insights.
Yes I believe it might be really interesting in having a more technical note on that (or perhaps a series of more technical notes on different topics).
I tried to replicate it by simply reversing standard formulas but I get much more symmetrical results in terms of implied probabilities (using SPX option @ 17/12/21, 17% <-20% and 16.47% for >20%).
Thanks again, Alfonso. That is definitely something to get started, but a technical post with more details would be appreciated. To be honest, I'd take some informal notes. :-)
Hi Alfonso, your website is teaching me a lot- thank you! Does the pricing for long term UST yields basically tend to follow the cumulative Forward OIS rates for the corresponding time period e.g. UST 10y is basically al the forward OIS rates from 0 to 10 years? Thanks!
Yes, plus the term premium which is a theoretical premium an investor would require to hold long-term bonds rather than buying and rolling short-end bonds over and over.
You are very informative!! I believe there are other factors that could be considered to derive a better result
Very interesting analysis and well written
It would be great if you could share how you calculate the implied probabilities in more detail as suggested in other comments below. Keep up the good work! Really enjoy your substack
Thanks! I'll think about publishing a more technical note.
Hey Alf. Love your works. The clarity of your writing is superb.
Beeing a Montary/ finatial noob I am learning a great deal.
In some of the articles you are using specific terms that are domain specific.
An example here is OIS and that great you explained the term.
But still there are plenty of such examples that will come up again and again.
ie. alpha beta gamma delta ffr ois …
Some the perspective someone that is learning and do this on an adhoc basis.
I’ll most likely forget the meaning of terms terms a couple of times before they sink in.
So it probably would benefit me the reader and you as the author so that you won’t have to repeat these explanations.
If this makes sense to you think about adding a “glossary” article / page with proper href references for u to use in other articles
Ie.
Tmc.substack.com/terms/#ois
Best of luck on your efforts.
regards Maciej
Thanks Alfonso, really good post! May I ask which Bloomberg tickers you use for the forward OIS rates, please?
The best way to build the forward OIS curve is to go on ICVS, select the USD OIS and play with the curve analysis part.
Hi Alfonso, very helpful post indeed. Are you using 1M forward rates here such as USD OIS Fwd Swp 6Y1M etc? Such as S0042FS 6Y1M BLC Curncy ticker in Bloomberg would allow to see history of it in addition to ICVS? Thanks
Ah perfect, got it - thanks for the response.
Enjoyed reading all your articles Alonso! Given that the option-implied probabilities were backed out based on the cost of various spreads, would the sum of the option-implied probabilities at all strikes add up to 1?
Yep, they should
Thank you for the great post. Wondering if you could share the method used in calculating the implied probability. It would be great if you could shed some light on whether such method can be used effectively in the FX market.
Hi Hafiz. I gave some hints to Paul, see my comments to his reply. I'm thinking about writing a more technical post in the future describing the process in details. Yes, I am pretty sure you can do the same with the FX market as well.
Thanks Alfonso! Been a subscriber for more than a week now and I have read all your posts! Keep up the great work!
Wonderful article Alonso. One question and one conment.
You use the option market for equities but not for rates ( where you use OIS). Why?
The problem with using options for equities is that will introduce bias in your estimates. A delta one long position hedged with puts is tax advantaged relative to a sale transaction that generates ( taxable) capital gains.
For rates, I am using OIS forward rates because they are pretty liquid also for maturities down the road. I am not sure I can say the same for Fed Funds or Eurodollar future options - sometimes you don't even have the underlying Eurodollar future contract so many years down the road.
You could use short dated options on deferred contracts. They are very liquid.
For the equity market I use a projection of ERP ( a 3 stage dividend discount model based Equity Risk Premium). That would suggest there is at least another 5-10% to go even if you allow long rates to rise 50 bps from here.
Absolute gold, thanks Alfonso. You said you wouldn't bore us, but actually I'm really interested in calculating option-implied probabilities like you did to produce your chart. Please could you provide a pointer to get started? Thanks again!
Hi Paul, very kind words. Thanks!
You can do this in several ways: one idea would be to use call spreads or digital options for different strikes and see how much you have to pay for a payoff within a certain boundary. That gives you an idea of how much probability the option market is assigning to a certain outcome within a range of strikes.
There are other ways as well: maybe I should write a more technical post about this once.
Hi Alfonso,
first of all, it is a real pleasure to read your posts: awesome insights.
Yes I believe it might be really interesting in having a more technical note on that (or perhaps a series of more technical notes on different topics).
I tried to replicate it by simply reversing standard formulas but I get much more symmetrical results in terms of implied probabilities (using SPX option @ 17/12/21, 17% <-20% and 16.47% for >20%).
Tks a lot
Thanks again, Alfonso. That is definitely something to get started, but a technical post with more details would be appreciated. To be honest, I'd take some informal notes. :-)