The Goal is: To design a program based on following well-established trendlines to achieve consistent and predictable profits. I am the visionary/thinker behind this, You are the tech-savy programmer. --- The program will function as so: The program will graph and go through hundreds of futures products, forex, and stocks. The program will draw trendlines, connecting the lows or highs of the chart (based on if we are bearish or bullish that product). It will only create trendlines with at least 4 points (or more- we can adjust this) and these points must be within 1%* of the trendline (the % is based of the total vertical increase/decrease of the product within the trendline period within which we are working). *(again we can adjust this/test different settings) -At this point, when a certain product nears it's support or resistance trendline again (for the 5th time), within 1% of the chart, the program will execute an order to enter that position. See the attached picture which illustrates this. ---- -The basis of this is the likelihood that the trend will continue- and if so we profit. To limit losses, the program will automatically close a position if it falls just 1% to the opposite side of the trendline (the direction we don't want it to go in) (thus limiting the loss to 2% of the trendline (note not 2% of total capital) which ensures very small risks. -We will only pick trends with at least a 25% slope in the sake of faster returns (as a 3% slope for example can be a waste of committed capital- better spent in another position) (we can again change slope requirements as we test). --- Regarding the attached example: The attached example illustrates a trendline of HLF (Herbalife) The trend began at price $24 and our entry point (the 5th bounce off the trendline) is at $31.50 Maximum loss potential: Right away, to determine our maximum loss, it's 2% of the price move of the trend so far, which is: $31.50 (our entry point) - $24 (trendline start point) = $7.5 2% of that is $0.15 So our maximum loss would be $0.15 on the trade, which decreases second by second as the trend continues. -- As for the gain: Entering at $31.50, we held our position until the trend ended, an exit point of $37.20 The total gain in this trendline was $5.70. -------- What we need: -Your ability to design a program to chart prices, and recognize trends based off the criteria above. -the program must then be able to execute buy and sell order - or at least simulate buy and sell orders while tracking profit and loss. -I need to be able to test the program and tweak settings to find the most profitable setting, after seeing consistent results a we can both use the program to our achieve own profits. -There are plenty of places for us to get free data feeds to input in the program and test this/tweak it to become consistently profitable. We will test it with a test account first - so no financial risk on either party- until we are ready to put it into play with real money.
don't know about current tools,but i'm pretty sure you can do this in wealth lab. i mean-automatically scan stocks,draw trendlines,backtest etc. http://www.wealth-lab.com/ you can probably find old version for free i mean-there is no need to invent and code this whole thing from scratch. and it's probably been done 1000 times already.
This is ultra basic, tradestation easy language stuff, why do you need a "programmer" ? Heck even a caveman could do it. Not to mention your ideas are out of the 1960's---- nothing visionary or revolutionary. Why not do some research before posting? surf
Just run a linear regression through the points and you'll get a trend line. You can find open source or free libraries implementing that in C/C++.
As a developer and systems operator here is a quote that I consider before each project: "Success in the markets is not about instinct, divine inspiration or spontaneous intellectual combustion. It is about intelligent data processing, sound method and the management of risk and resource that is both effective and adaptive to change." Today there are smart technologies available (smart data mining, neural networks, genetic programming & non-linear optimizations) that guide missiles against smart targets in 3 dimensions at mach 3. Certainly such technologies, when properly applied, can give one a brief look at the trajectory of the S&P. The technical part (programming/optimization/test trading) of system development is very inexpensive when compared to potential returns. What is rare and what adds the real value is the base concept. Today technical ability is a commodity and on the world market, is very inexpensive. The issue in most cases is that those few that really understand market dynamics don't understand the technologies that are available to help them gain greater profit from that understanding. With excess development capacity I am always on the lookout for fresh new concepts, have done co-development before and am looking for more. As one would expect I pass on most of what I am shown for co-development - but always still looking. This is one of those where the OP has no firm footing in either concept (market dynamics) or the technologies available to develop it. Jack
That was the scariest statement so far from the OP. That is what NN and GP do....optimize to the point of curve-fitting the strategy. That must be avoided at all costs. The backwards and forward testing methodology wasn't even mentioned.
My vote for the scariest statement would be any of his sentences containing the word 'trendline'. I can't think of a slower, more useless method of trend following than trendlines. Like surf said, right out of the Sixties.
I'm still looking for partner in this and I believe my methodology strongly-- Don't mind the negativity of others- their responses are a reflection of their own success or lack there of, As for myself- I've witnessed trendlines plenty- call it old school if you will-- but the principles stay true today as ever before- Price trend- that's it. Volume will indicate pressure but is not even necessary to measure, the trend is by far my primary concern and it is what will be the bread winner. I am not a programmer, again my focus is on the function of the system -- This is not rocket science, I implore any of you who have the resources or ability to do this to contact me, lets make it a win-win and build a long term relationship and work toward our mutual benefit. Cheers~