statsign

He is extremely brilliant. I can say with complete humility that the economy of the USA can be modeled and predicted. The requirements of doing this are asking the right questions. The Fed has zero interest in how the economy works for anyone other than the owners of the Fed. The assumption that they do is at the heart of the article's folly. And, "Several equations cannot capture the simultaneous reactions of six billion people around the world to a million different economic events." stated is a continuation of folly, world based instead of USA based. The modern acceptance of free trade that has historically always destroyed the economy that began on top. Many variables there are, but that isn't chaotic. Equations from chaos theory, and more so - non-linear dynamics, are extremely useful in finding root causes, but once found the end result is the knowledge that the proper questions lead one to a linear system of accurate prediction.

An index future doesn't have to follow the index because it is a bet on the future and when it doesn't it can be an indicator. Trading index futures has some big advantages, namely the liquidity you seek. It is a whole lot easier to get a few dimes on 100,000 shares than a dollar on 10,000 - so it should be obvious that the biggest edge a day trader can have is money. One of the main reasons index future traders fail is trying to stay inside the time limits for low margins. For months all moves happened outside those hours. And again, it very often just comes down to money. Trying to make $50 on a $65,000 future doesn't make good sense without a serious edge. That type of edge is rare across the board - regardless of player size. It is achievable though. My opinion for $100,000 would be to work a system with mostly winners on big stocks for dimes on 1000 shares until a few hundred a day is steadily pulled in, or better use their options instead if the system is comfortable.

I do understand, but I think this is good to have post on. I'm not the best writer for sure. When I say call the function, it is because it isn't inline and code has to move to the pointer for it. Certainly today, the delay for that is giga, but on ES ticks with 8000 lines? (rhetorical) I understand what you mean - keyword, so I explain myself, boolean call to math to boolean - wrapped in a keyword. Anyways, speed is king and if auto daytrading definitely go for overkill on processor speed and memory and internet speed.

yaeh, I'm always thinking strategy coding, but for indicators using functions is fine. In a strat calling functions takes time, crosses is a boolean that calls the math to set the boolean - 4 steps instead of one

Mom > 0.5 and Mom <= 0.5 Mom < -0.5 and Mom >= -0.5 are both impossible - you need to decrement 1 bar Mom > 0.5 and Mom [1]<= 0.5 Mom < -0.5 and Mom[1] >= -0.5

This trendline indicator was posted on TS support a few years ago. probably in EL Library

My apologies JKLM, I thought I was subscribed to thread but received no indication of your post. Forgot I posted here or wouldn't have opened it now. docount stops the tradecount from continuously incrementing when in a trade

Code has to run to do calculations for auto trade. There are brokers that will run your code if you are comfortable with it. MovAvg xovers are not so good but entries are not so important. Exits are important and ATR is not good at all for making money. Anyways, I won't give anything away which I know is frustrating. But I am a master statistician and a whole lot of things the rags and gurus tell you to chase are losers. I will tell you that price is everything - and I understand that ATR is based on price, but it already happened and has zero predictive usefulness. One thing I've heard so many preach over the decades is to keep it simple. I'll tell you - keep it as simple as is profitable, and there is absolutely nothing simple about it.

I keep to stats - growth is just a percentage of capital if system says setup wins > 85% of the time it will risk 30% > 70% < 85% risk 20% > 50% < 70% risk 10% of course profit is all about exits

EasyTrader_I , I code for speed, a counter in a strategy is faster, I don't use any functions in a stategy

sure - declare variables wrap your order code inside of - if tradecount < mynumber then begin end; reset a trade count every day - if d <> d[1] then tradecount = 0; make a trade count - if marketposition <> 0 and docount = 0 then begin docount = 1; tradecount = tradecount + 1; end; if marketposition = 0 then docount = 0;

8 bars ago, 5 bars ago, timing has to be awful

I have a preference for booleans, just how my mind works, but doesn't have to be long example input: targetv(2); if marketposition = 1 then begin mytarget = entryprice + targetv; if C = (entryprice + targetv * .5) then mybreakeven = 1; if C = mytarget then sell next bar market; if mybreakeven = 1 and C = entryprice then sell next bar market; end; if marketposition = 0 then mybreakeven = 0;

right, which is why I don't use functions at all in strategies - thus all orders are buy or sell or short or cover