419 reputation
23
bio website
location London, United Kingdom
age
visits member for 3 years, 4 months
seen Oct 1 '11 at 12:11

Aug
26
comment Can one use options on Treasury futures to hedge a portfolio?
The Black-Scholes delta would be wrong as it assumes non-stochastic interest rates. The Black model would be a better assumption (even that doesn't take into account the volatility term-structure). Also delta*duration is a naive estimate as you can't really decouple the delta of a bond/bond-future option from the duration of the underlying.
Aug
23
comment The T+H Problem in Factor model forecasts
You are assuming that you are still using your T+H model. What I have suggested is a "local" volatility model that gets recalibrated with each passing day. If you can add the factor coefficients to the set of variables (which includes the historical weights as described above) to be optimized such that the "error" is minimized, then you can do this on a daily basis using PCA (or any other method you prefer - PCA would be based on a covariance matrix which varies daily).
Aug
23
comment The T+H Problem in Factor model forecasts
Point (b) would help if you have a sliding T window. Or are you telling me that once your model makes a prediction about time T+H, it never adjusts it? (implausible considering H is 1 year!). As far as realizing (c) goes, something like taking 10 points in your T window, giving them weights w_i and then solving for the optimal set of weights that minimizes the difference between today's observed/predicted values. Once you have the 10 optimal weights, use spline-interpolation for intermediate points. This helps you get an optimal "weight" curve for each of your historical points.
Aug
23
comment What type of investor is willing to be short gamma?
@glyphard - slightly incorrect, you can be (fairly) gamma-neutral while having a net volatility position - eg a straddle/strangle. Also, a short put is short gamma but long underlying.
Aug
22
comment Is there any gamma in basis (i.e., floating for floating) interest rates swaps?
in general, yes - there will be gamma even then
Aug
19
comment Proof that you cannot beat a random walk
@sheegaon you can't "trade" a single asset - even if it were a single stock you'd need cash - which is another asset
Aug
18
comment How do I replicate John Hussman's recession forecasting methodology?
Also, regarding the use of the 2y10y spread (or equivalently the 2s10s on the swap-curve): (a) the 3m point on the curve is not usually determined by the swaps market but by money-markets and primary Treasury supply - swaps market have more liquidity at the 2y point and are more liquid. More often than not, the 3m is "implied" (via no-arbitrage arguments) or interpolated from the swap curve rather than the other way round. (b) if a recession were to hit now - you wouldn't see much movement in the 3m rate, however you would see significant movement in the 2y10y spread.
Aug
18
comment How do I replicate John Hussman's recession forecasting methodology?
@Zach: CMT = constant maturity treasuries
Aug
18
comment Proof that you cannot beat a random walk
@Bootvis - this argument does not take into account dynamic replicability of a position. I can have a risk-free bond and a stock - both having iid returns/price processes. I can create an option to exploit the volatility characteristics of the processes. See my answer below.
Aug
8
comment How do I replicate John Hussman's recession forecasting methodology?
should have been "4." for the last item in the list :)