Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Sorry the other link didn't open. The simplest way to find it is to search foxforce5. Two articles should come up.
Regarding your criticisms, at least some are debatable but I really appreciate the effort.
That article was based on price being above or below an indexes EMA. It didn't include Specter or Foxforce5.
Theoretically buying at the close is OK and possible. You know what the signal is when the market closes and should be able to buy at or close that price (even in the aftermarket). I'll do a comparison with buying at the open the next day, it makes little difference in the long run I think (favors the close if it makes money). Buying at the close is widely used in strategy development -you either buy at the open or buy at the close. So your criticism might apply to anyone here who develops strategies. Also buying at the close is a legitimate order type (I've never used it).
Your comments about lack of data are not quite fair. I'm writing an article and am concerned with making it readable, so I'm not dumping all the crap I came up with but selecting stuff I think the reader might find interesting. As you point out, I probably failed on both counts but that wasn't related to making silly data mistakes. Also I'm analyzing The Dow Indexes so your comment about indexes isn't clear.
Yes, the peak and troughs were known beforehand, because the guy I cited came up with the dates which was the first thing I showed. I even mentioned that I didn't want to go to the effort of figuring them out myself.
I showed results over 90 years so any backtesting issues mystify me. I wasn't trying to prove the strategies were good, just wanted to show the return over various time periods. They weren't really recommendations. Most of my issues are communicating with humans.
Thanks again...I got an idea reading your last message that should make this easier to understand. What blew my mind is that making the change made the engine incredibly faster and more reliable. I've been worried about proving my calculations are right - I'm about 95% sure. This new technique will make that easier to prove (or not). I can set up a new thread with that stuff, seems I've put this one off course a bit.
I'll buy you a Rolex or something if this gets monetized.
Can you help answer these questions from other members on NexusFi?
MOC order that what its called, the best order for illiquid stuff as you get to participate in the end auction where there is hopefully more liquidity and fair value.
But I think you don't understand my comment on first future peak observation. You suggest a weighted price: (low+high+close+close)/5 and an enter on crossover with the EMA of this price:. How can you know if this weighted crossed over if you do not have the close yet? Maybe eyeball 15 min before close if the signal has crossed and enter on MOC? Agree this will work sometimes but you will also miss entries and you will get in when it hasn't crossed. Especially at close you get these forced flows from leveraged ETF's which can drive price really up or down. You typically always want to be on the conservative side with an analysis.
The future peeking with these periods (2 months) will have a much bigger impact on the results of the analysis as the latter one. You really should redo that part if you want to take your result seriously.
About my not fair comment, the amount of observations. I have said that this research on all the index historical constituence was not really in scope of the article. Its just something what I would do to have more data, but is hard and cumbersome. So imo nothing unfair about that. But you have understandable decided not to go that path to keep things short and simple, than I would expect you to mention something about lack of observations in your conclusion.
Forgot one comment, you base signals and performance on an index. Signal is ok but performance is a problem because an index is not a tradable product. We know in market stress, these sector ETF's can deviate from the underlying.
(Open + High + Low + Close + Close) * .2 - Dividing by 5 is impolite, much tougher for the computer than multiplication. I've used that for 20 years, it makes no important difference in the strategy results, don't think it makes them worse. I like to use something different than other people, so I invented that, weighted price, typical price or closing price there won't be much difference. The weighting gives you a chance to use a 1 day moving average over the close, spent quite a bit of time with those also but not all that promising.
I wanted to make the articles so the system can be reengineered if someone (presumably someone pretty good) wants to. I'm more interested in research than trading profits.
You want to get an idea how a strategy works, if its dependent on nickels and dimes to make profits its dubious anyway.
I had to do the indexes to go back 90 years. I have daily data from Norgate for 20 years on listed stocks. 30 years is more money and its not often you want to go deeper than 10 years. So only had the Dow Jones data with suitable detail.
I started looking at enhancing strategy profit with Specter, but only after I saw that that outperformed vanilla moving averages. Foxforce5 is a new design of the strategy logic engine. That makes complex strategies simple to analyze, it merges above/below and up/down states. Those two elements together are very strong and that's the returns on the tables I've published here. I spent a year full time on building accounting logic before getting to the playable strategies part, but the base part of the application is multiple securities excel vba handling was developed from the mid 90s on as a hobby.
If you are a hobby researcher not and having skin in the game is imo perfectly ok. But than I would advice you to be a bit more humble (esp early posts). Other people here on this forum like me do this for a living. They feel the pain if they have money on the line based on flawed assumptions.
Still looking for a someone else to post data. I'll be humble when I see how the people play strategies for a living. More than a few posts on here seem to be from people who are drunk.
Best way to find out is to get into the game yourself. This way you will learn fast. Then you know like you never get the opening price as assumed in backtests or you know what lack of liquidity, margin, slippage, tradingcosts will mean for your PnL.
I've been in the game since 1980 when I started to work on Wall Street. You're just taking a comment I made about my motivation and ludicrously misinterpretting it to imply I don't know what I'm doing. That may be true, but I know that applies to the vast majority, if not all the people on here.
Not a single person on here says exactly what they do and how much money they make, you're pretending you know what you're doing. I'll give you the benefit of the pretty serious doubt, but don't go insulting my intelligence, I didn't insult any of your comments.
I'd be surprised if there are 10 people here in this forum that make money with strategy based trading. There are probably others that somehow know how to trade by experience and can eke out a living, but don't understand how they do it exactly.
Whatever this is, it's a hang out not much different than a bar except its by computer.
Sorry if you feel offended by my comments, was not my goal here. And yes I make assumptions/generalizations on conversations I have with people. Can't help that, somehow you ended up in my starting strategy researcher/developer department. And tanks that you give me the benefit of the doubt altough I'm not here to pretent to be a know it all. But I'm curious and like think of myself as a generalist, eager to learn all aspects of trading and not getting stuck in only one little niche thingy trading only one or two instruments. Which can make money btw, just nothing for me.
My goal is to share my experience with other especially starting algo traders, who most of the times are missing out on reliable information and/or have over optimistic expectations or flawed assumptions. With hindsight I can say that I really would have loved that support, would have saved me a lot of money, I've been doing a lot of stupid stuff I now see others beginning traders do.
Only thing is that starting traders imo will not take any advice, maybe that would the same in my case. People that start think they have a clear goal. They want that one thing, that indicator, that strategy, something totally irellevant most of the times. Guess it just like the law of gravity every retailer start trading will have to fall and blow up some accounts to come back to earth. So might as well spend my time elsewere.
