Ya I guess I am a non trader person. Your right, I wouldnât trust me either. OK. Taking the advice I ve seen in so many threads and PMs. Creating my own account w 10K. Will start trading after the fiscal cliff and the Mayn calendar thing is settled. I like the Journals forum so once I get going I will try to post results there as 2013 unfolds.
I think need forward testing. No point huggin to all the backdated data and applying a simply buy and hold algo and claim u are making 30% / yr.
Need to ask, what is meant by forward testing? Do you mean applying to live data? Exchange activity for each stock as recorded by my feed is processed "live". I post my buy minutes, hours, or days before the sell. As for the claim, I get it. Put up or shut-up. I will start a journal next month for 2013. Previous test was w Scottrade. Will see if there is some kind of report screenshot, etc to post. Not going to post an xls which will be called fake. As a side note, I am turning off the twitter feed as it met with 0 interest and one complaint; go figure. Who would take the advice of a dumb a** non trader anyways If anyone did find it interesting, let me know, and I can turn it back on. Ever notice Jack D has a very distinctive flavor.
doesn't really matter if you post it up here or not. Forward data means u can anticipate and act accordingly, either stopping further losses, and capturing more Alpha. End of day u dun answer to elite traders forum. U answer to ur own acct. I've been doing this forward testing for 2-3 years, running 4-5 portfolios. What you are doing is nothing fantastic yet.
OK, Humbled. One last try. From your past posts looks like ur sh*ts together. Could I ask how long it took to hit big numbers? I only ask as I look at my own forecasts. I am working in a vacuum here.
what are your risk stats, Sharpe, max drawdown, etc.? 30-40% in a vacuum is meaningless without some understanding of risk. It sounds like you have a solid technical background but weak in statistics/risk management. As others have advised you need to trade it to get real results. I think you will find there is a lot of random unanticipated crap that will happen, and errors tend to blow you up instead of help you. A forward test is when you take a backtested model and test it on live data without changing anything about the model.
This is all sounding so....phony. You are beating 90% of all money managers with those kinds of returns. And wtf are you doing in today's low-paying IT business anyway if you've got that kind of systems trading talent ?
Hold on. Never claimed to have any trading talent and wouldn't even try to attempt the things done in some of these forums. IT pays good, family is well fed. No complaints here. I simply got excited by the model results and wanted to chat about it like the dumb a** I am apparently. Good luck and Happy Holidays.
OK found a report on Scottrade that shows the first test. Started Feb 2010 with 5K cash account. No losses. return calc=(Value of investment at the end-Value of investment at beginning) + Dividends / Value of investment at beginning (6904 - 5000)/5000 = 38% in 15 months. I did lie I thought it was 12 months. Missed lots of trades due to 3 day waiting period and busy work sched. 38% of what I got is still nothing. Second test started June 2011 - Oct 2011 was with Prop firm. Not enough model activity for the effort. I was too timid to take on same day leverage. Didn't take series 56, so stopped. Since Oct 2011, new release of code.....got excited came here, was told to F off. Third test will be Jan with 10K.
Very odd. Polarbear in the journal forum, trader p/l ,is getting 850% pct return, 3.5 Sharpe, max DD of 60%. And I get crap for saying 30-40%? Yikes.