Search results

Wtf? I never claimed to have traded my model yet. I like to fully and completely understand (to the extent that this type of thing can be understood) what I'm doing before I put money on the table. Also, I thoroughly enjoy using profanity and my 'credibility' doesn't affect how I make money...

Thanks, NF is definately a great board, lots of people there are the real-deal.

I'm offering inspirition to those who want to do so. And I also get a kick out of ruffling your feathers.

I don't care if mandelbrot says prices are unpredictable or not, I doubt he has studied super high frequency data very much. In fact, in this Article in wired he suggested that acquiring high frequency data would cost "20 million dollars"...

Well, my previous model was not making any directional forecasts at all and I felt that I probably could be doing that as well and combine them. So I opted to do some more development before going live. I'm happy with the price forecasts now, and as you said, it is one thing to have forecasts...

Yeah I don't understand any of it, I type in random shit and somehow these accurate price predictions coming flying out of my computer. I decline your challenge, go do your own work, I'm not giving out any freebies. Also, taken's thorem cannot be applied to the raw price series, but I...

I'll be nice this time. 1st differencing loses all trend information, that is why it is a bad transform. It is typically done because it is easier to work with stationary data rather than non-stationary prices.

No offsense, but dont read the post if you dont like my form. Predicting EOD data is insane, like forecasting the weather a year in advance, unless it is something semi-periodic like El-Nino.

Where did you get this idea? "Gaussian Self-Affinity and Fractals: Globality, the Earth, 1/F Noise, and R/S" Fractal/chaotic processes are clearly predictible in the short term if you have enough data, find the proper embedding and then find the appropriate lower-dimensional phase space...

I don't follow. A non-forecasting type of strategy would be basket spread trading where you dont bet on the individual price movement of a single holding in the basket.

I don't have any futures feeds yet, and I still need to port this over to my realtime platform from matlab..

Yah of course its useful.. when you get down to the highest frequency of raw ticks, it's not that useful because you are then trying to forecast the mostly-unpredicble actions of a few people/bots/whatever. The aggregate behavior us much easier to see. The weatherman doesn't tell you the...

http://en.wikipedia.org/wiki/Stream_Control_Transmission_Protocol - multi-homed - connection-oriented similiar to TCP - automatic rebroadcast/assembly of dropped packets - multiple streams within one connection - smarter routing - handles congestion more easily than udp lots of others...

Yeah, well it's definately good news, hopeuflly it'll develop into something. I wish people would start looking into STCP rather than UDP for multicast dissemenation. http://www.csm.ornl.gov/~dunigan/net100/sctp.html

Without giving away specifics, do you forecast price changes, price direction, horizons, etc? I used to think this was impractical but I've changed my mind. Taking first differences of prices is a very high frequency transform which removes lots of valuable information, as a test I used a...

Any ideas how long this compression takes? Still don't think it can be specialized binary protocols..

This explains my philosophy the best. "One should respect public opinion insofar as is necessary to avoid starvation and keep out of prison, but anything that goes beyond this is voluntary submission to an unnecessary tyranny." --Bertrand Russell So, I don't mind pissing people off...

Well, really obvious things like the QQQQ and the nasdaq100 futures are cointegrated by design, but you really cant make any money there cause the market is so effecient.. just gotta find em, you can do simply regressions and tests in excel, the yats.com linked posted above is good.

When you get into timeframes of days and weeks you probably only want to use cointegration based pairs as correlation is only useful for short timescales.

I'm not tied up in anything. And QQQQ/SPY is a very tradeable pair, even with the divergence, maybe you just aren't smart enough to see the chaos.