| bio | website | |
|---|---|---|
| location | ||
| age | ||
| visits | member for | 1 year, 7 months |
| seen | 24 mins ago | |
| stats | profile views | 11 |
|
Mar 12 |
awarded | Scholar |
|
Mar 12 |
accepted | Pairs trade CDS contracts using cointegration |
|
Mar 12 |
comment |
Pairs trade CDS contracts using cointegration Ok: having bought the CDS, I sell it before having to make the next quarterly payment, and vice versa, having sold the CDS, I'd sell it before—when? Default? (might be hard). I guess as a seller I would like to receive the quarterly payment. |
|
Mar 12 |
comment |
Pairs trade CDS contracts using cointegration With a position in the underlying bond you mean? |
|
Mar 12 |
asked | Pairs trade CDS contracts using cointegration |
|
Mar 9 |
comment |
Pairs trading: Question on non-negative profits, size of the positions and trading signals I don't see why you would want to go long-short in equal amounts: the cointegrating relationship says that $X_t - bY_t = a + \epsilon_t$ is stationary. In the case of just two variables, the cointegrating relationship (cointegrating vector) is unique up to a scalar (i.e. $\lambda X_t - \lambda b Y_t = \lambda a + \lambda \epsilon_t$ is also stationary). This means that unless $b=1$, you DON'T want to go long/short in equal amounts; rather for every dollar you go long (short) in $X_t$, you want to go short (long) $b$ dollars in $Y_t$, since otherwise your position will be nonstationary. |
|
Feb 9 |
comment |
R ARMA-GARCH rugarch package doesn't always converge Have you tried several of the different optimization methods that rugarch offers, or just the default? |
|
Oct 28 |
awarded | Disciplined |
|
Oct 27 |
awarded | Editor |
|
Oct 26 |
awarded | Student |