yeah,
but do ya work with Neural Networks? I just happened to read a post from an ET illuminatus saying that kwants also do NN.
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.
Anti-Jack Trading
- Thread starter hypostomus
- Start date
- Status
- Not open for further replies.
but do ya work with Neural Networks? I just happened to read a post from an ET illuminatus saying that kwants also do NN.
Nononsense, I once had a fairly accurate (and probably totally spurious) linear estimator of the S&P 500 based on the yield, the 90 day T-Bill rate, and M3 (you remember what THAT was, haha!). I got there by correlating the S&P with every bit of data I could find, like the size of the Ukrainian potato harvest and the price of rubbers at the corner gas station. Only the aforementioned three had convincingly high correlation coefficients.
So when I got interested in NN's (any correlation THERE?), the first thing I read is that they run cross-correlations on everything in sight to figure out what to plug into the NN. After I picked myself up off of floor and changed my shorts, I kept reading. Oh! To first approximation it's a linear equation. Duh! Oh! It might go nonlinear. Uh huh. Finally, it might become multiplicative. Oh, yeah, the price of rubbers times the potato harvest. Bye bye!
Nowadays anyone who can research and READ (excluding most of ET) can get the answer from Katz and McCormick's awfully-titled "The Encyclopedia of Trading Strategies'. It ain't a pretty conclusion, glad I stuck with Kalman filtering. Joe.
So when I got interested in NN's (any correlation THERE?), the first thing I read is that they run cross-correlations on everything in sight to figure out what to plug into the NN. After I picked myself up off of floor and changed my shorts, I kept reading. Oh! To first approximation it's a linear equation. Duh! Oh! It might go nonlinear. Uh huh. Finally, it might become multiplicative. Oh, yeah, the price of rubbers times the potato harvest. Bye bye!
Nowadays anyone who can research and READ (excluding most of ET) can get the answer from Katz and McCormick's awfully-titled "The Encyclopedia of Trading Strategies'. It ain't a pretty conclusion, glad I stuck with Kalman filtering. Joe.
Pretty funny. This evening at 19:41:21 ET the B/A for NQ went to 1568/1569.50 (6 tick spread) and three cars actually popped at the bid. I don't recall ever seeing the spread widen that much after hours. Someone somewhere was not paying attention! How can you miss that?
Even saw it widen a few times to two ticks intraday. So there's hope for the guy who wants to trade index futures inside the spread!
Even saw it widen a few times to two ticks intraday. So there's hope for the guy who wants to trade index futures inside the spread!
I forgot to mention, as an anti-Jack rant, that the other day in a tight consolidation I watched 1000 long market order cars flung at an inside market showing a depth of 4. The ask didn't budge. So don't tell me the DOM means a shit!
Quote from hypostomus:
I forgot to mention, as an anti-Jack rant, that the other day in a tight consolidation I watched 1000 long market order cars flung at an inside market showing a depth of 4. The ask didn't budge. So don't tell me the DOM means a shit!
Hype: are you saying 4 lots were offered at say price X and it never moved while 1000 lots went over the tape at X ?
Iceberg limit order at the ask ?
DOM info is like other info: dead right and dead wrong. Of course Jackers know exactly when is which and thats why they are all rich as croesus.
CL. Exactement.
- Status
- Not open for further replies.