Disliked{quote} Is there no software that does it? Playing with numbers can't be that hard. Yes, it's for ALL major FX pairs and one derivative designed especially for the EURAUD and EURNZDIgnored
Staying in my lane...
Another Day, Another Chair, Another Day in the Office 198 replies
Systematic trading 1,566 replies
Nonparametric Association Measures in Systematic Trading 57 replies
A Systematic Approach to Markets 68 replies
Systematic Trading vs. Discretionary Trading 14 replies
Disliked{quote} Is there no software that does it? Playing with numbers can't be that hard. Yes, it's for ALL major FX pairs and one derivative designed especially for the EURAUD and EURNZDIgnored
DislikedAnyone here trades stat arb models instead of indicator based systems? I have developed one (out of curiosity that involves basically the DOLLAR (Index)) and it works extremely well.Ignored
DislikedWhat was the longest period your bot was running successfully without touching or tweaking it? Anyone?Ignored
Disliked{quote} can you shed some light here. are you referring to basket formation with weights etc? is it momentum or mean reversion?Ignored
DislikedUSDJPY 0H-2H BREAKOUT RESULTS: Profit factor = 1.22 avgPips = 2.92 #trades = 2681 2009-2019 transactionCosts = NO Timeframe = 1min Probably it needs some filter to overcome the transaction costs and make it profitable, but there is some edge as Alpha pointed out {image}Ignored
Disliked{quote} I am also hoping anyone can prove me (or CP) wrong when I say "Any strategy that trades every buy and sell signal on a price/time series will fail in producing consistent returns". I tried to prove this wrong and I could not. I think CP is right. There is something about directional bias that makes a strategy win or lose in the long run.Ignored
1: The problem with this attempt is that (from my experience) you'll always deteriorate the parameter r with increasing R:R. Thus, in the long run you'll alwys lose with a strategy like the one shown in the link.
2: I've only partly surveyed that strategy (there's still a lot of work to do). No, I haven't found an edge yet. Assume that your statement above is true: The use of 1 indicator for entries and exits will always yield losses in the long term. Now let's assume that you use 2 indicators, one for entries (e.g. determining momentum and enter if you see strong momentum) and another for exits (e.g. another momentum indicator that tells you that momentum flattens). If the signals of both indicators are still highly correlated, you would still lose. But what would happen if the correlation of the indicators is almost 0? How could it be established? Maybe using a lagging and a less lagging indicator.
Disliked{quote} I have already tried this strategy what you posted earlier. It's a loser on NQ as after 2 hours of NYSE, most likely price reverses (unless on very strong trending days like we had in january) ....Ignored
Disliked{quote}
Ok, time for some theoretical nonsense:
- Although alpha doesn't like it, there needs to be an edge in trading. What does it mean to have an edge? It means that the strategy needs to make you winning (P) more than you lose (L). If your win rate is r, any successful strategy should fulfill the inequality L < r × P.
- There are only a few scenario I could think of how to make this work:
- Add to winners to increase the potential profit.
- Enter with one indicator and exit with the signals of another indicator.
...
Ignored
Disliked{quote} Dude - I am only focused on NQ and nothing else and only during NYSE session and only with 2019 data. NQ has a completely different footprint. It can work but it will also cause DD using rangy days (similar to any breakout style) Is there any way you can load tick data of NQ in your backtester? I could use some help as Sierra limits to 90 days max. I can no longer test with January tick data now :-(Ignored
Disliked{quote} Vee, I am not sure what you are trying to get at. this strategy that I presented; has an edge on USDJPY. I re-run the backtest; and have similar results to DavidRP; with extremely simple rules: - backtest date range: 2011 to present day; using tick data (raw bid/ask) - buy when the price breakout from the first 2 hours range (GMT 00:00 - GMT 02:00) (opposite for sell) - Exit on opposite signal - Exit at end of day (NY 16:45) I haven't used any stop loss, neither profit target, neither trailing stop. since the curve is positive; there is...Ignored
Disliked{quote} I hate to even mention this as it just sounds retarded but I will say it anyway (this is specific only to NQ and only during NYSE)....of all my testing rigor, the only 'approach' I have found till date feasible and close to my overall success criteria is to do the following Entry via brute force across multiple strategies specific to trending or ranging market. (I remain unsuccessful to devise a strategy that will work in both market conditions but I am getting there with the HHLLHLLH market structure!) enter with no SL and small TP like...Ignored
Disliked{quote} highlighting areas where human intervention will take place.. good luck if you started the equity curve where the blue line is...that's torture right there. You seriously think there's an edge here..that's almost 3 years of stagnant performance! {image}Ignored
Disliked{quote} Hilmy, that curve is the raw edge curve. you have to add some filters or exit techniques to improve it. in my real strat, i use multiple TP levels. that is where traders can add value. the edge existsIgnored
DislikedBy the way, I think the word "trading edge" is thrown around without fully understanding of what it means from investopedia: A trading edge is a technique, observation or approach that creates a cash advantage over other market players. It doesn’t have to be elaborate to fulfill its purpose; anything that adds a few points to the winning side of an equation builds an edge that lasts a lifetime. Don’t be frustrated if you haven’t found one yet because the majority of traders don’t even know it exists. It’s the primary reason why a few book excellent...Ignored
Disliked{quote} great discussion... exactly the type we should be having to grow our knowledge and skill in this space. I feel I struggle more than others because of the variety of channeled thought process I find myself in frequently. Not sure if it is a good thing or bad thing yet but I do find it extremely hard to convert a "raw edge" whatever that is... into a provable working edge by the so-called filtration process. In my experience the opposite has happened... the freaking edge (or the illusion of edge if you will) starts to DECAY the moment I introduce...Ignored