You are an excellent trader/communicator. The concepts in all your videos are really good --- I like the way you keep it simple and don't over-optimize. One small request; can you add realistic slippage and commission to make the equity curves more accurate/believable?
Thank you for your kind words. Will incorporate S&C in future videos. If you use average trade then you can be confident about your profits. I add S&C at the last stage of development, but I make sure that the average trade is at least 2 times S&C
@@StatOasis Ali, thank you. I am looking for the method you used in the stock selection illustrated in your video. Could you share that with me? Thanks so much. I appreciate your videos.
I just watched the video and honestly I don't remember the sorting because it is 1 year old, but from the video I think its based on capitalization. there is no sorting function in SQX 135 so probably I picked the famous names also 😁, SQX version 136 has sorting, but I didn't spend much time with it as it is still buggy.
Hi Oasis! I wonder is there any efficient way to algorithmically trade stocks inside and of Stocks and Share ISA for me here in the UK, none of the brokers offer sophisticated order management is my problem
I think tradestation is available in UK, if not then you need to run an algo platform and enter orders manually in interactive broker, there are also services that take manual orders and send it to IB
of course, but only testing will tell the truth. next version of SQX will be able to test single strategy on multiple symbols, so that will be a good start on stocks.
Ali, when trade a set of the underlying stocks instead of the ETF, did you divide the amount of capital you commit to each trade by the amount of stocks used in the portfolio? So if you capital at any given time is 10000 and you are applying this strategy to 10 stocks, you buy shared for 1000?
I really wish you would've showed a chart. All of don't have algo systems so we need to work the system manually but how can we do that if we can't see how everything is suppose to look? Can you keep that in mind next time?
you should not trade one strategy, there are many studies that shows the power of uncorrelated strategies in a portfolio, even 3 strategies will make a huge difference www.statoasis.com/post/the-holy-grail-by-ray-dalio There are strategies that are built like a portfolio, where they trade multiple assets using multiple factors, like TAA (Tactical Asset Allocations). those can be traded alone, there are many flavors of them if you search the term TAA. I currently have 8 of them in the VIP+ channel in my community go.statoasis.com/community
Could this be optimized for lower timeframes? strategyquant is a little to expensive for me. Also, how are you only getting an average trade of around $280, each contract is $25 a point and you are going in 100% with a 100k account. That around 180 contracts (assuming broker). So either its trading micro points or only trading 1 contract at a time.
Excellent idea! 1.Is this strategy only long position? 2.A quantitative strategy that has been extended to a basket of stocks, a good idea. I hope to share more stock strategies developed by SQX in order to increase diversification. If this strategy can be extended to the advanced strategy of searching for the most suitable stocks for thousands of stocks, and only using SQX, I would be very interested in this. 3.Regarding a portfolio of 12 stocks, should I use $10,000 or 120,000 as a starting capital? Obviously these two values are unreasonable, so how much should be more reasonable? I already understand the starting capital selection method of a single strategy, but in the portfolio? Is it also through Monte Carlo simulation to check the ruin probability?
@@StatOasis Thanks very much， i watched the video "What is the best Money Management method for a portfolio?" again, and I have a new question and leave a message under that video
@@StatOasis ok, I guess I should've asked: To get the historical data back to 1993, are you subscribed to the SQ data package or did you import it yourself from a different source?
I have SQX data, and Tradestation data, both go back to 1993, SQX intraday go back to 2009, but I don't remember which one is used in the video. FYI, if you want to subscribe to SQX data, use "StatOasis" coupon to get 23% off
If I understand the description correctly, you buy the dip on the 7th down day then sell when the asset hits a 7 day high, so it’s BTFD. How then are returns flat through all of 2008 and half of 2009? All BTFD strategies get slaughtered in bear markets because you’re continually buying back in on the way down
I can not replicate base strategy results on spy. I am bringing Tradestation data into sqx. I can go to 60 k max while you go upto 82 k. What data source are you using? I have triple checked my settings to match yours.
this one was using SQX data provider SPY day session. not matching my results is ok, but you should be very close, as even same symbol from two data providers will have different results but very close. check start/end date, variable shifts, daily session, etc.
Returns are actually a little bit higher if you include all delisted stocks. I chose this route so viewers can replicate the test, but to include more expensive data that include all history for all constituents and platform that can do the test will be hard to replicate for my audience
It’s a different profile, but correlated to RSI2. Which one is better will depend on what are trying to achieve in your portfolio. If you are looking for best returns on single instrument then it’s a lot easier to compare
I believe Norgate produces data that fixes the survivorship bias issue, possibly need to use it with amibroker as it`s integration with multicharts etc seems limited,,,would be interesting
You are an excellent trader/communicator. The concepts in all your videos are really good --- I like the way you keep it simple and don't over-optimize. One small request; can you add realistic slippage and commission to make the equity curves more accurate/believable?
Thank you for your kind words.
Will incorporate S&C in future videos. If you use average trade then you can be confident about your profits.
I add S&C at the last stage of development, but I make sure that the average trade is at least 2 times S&C
I am always glad I found your channel :) keep up with the good work
Thank you for your kind words 🙏
Ari, what method do you use for selecting the stocks to trade? And how often to you refresh your criteria?
Best way is to use ranking
@@StatOasis Thank you. May I ask what criteria you use to rank them?
there are many options, returns, pullbacks, volatility, etc.
@@StatOasis Ali, thank you. I am looking for the method you used in the stock selection illustrated in your video. Could you share that with me? Thanks so much. I appreciate your videos.
I just watched the video and honestly I don't remember the sorting because it is 1 year old, but from the video I think its based on capitalization. there is no sorting function in SQX 135 so probably I picked the famous names also 😁, SQX version 136 has sorting, but I didn't spend much time with it as it is still buggy.
Thanks for the video, one question, does a time stop work on strategies like this one?
exit after number of bars works very well with most strategies
looks amazing! Thanks for sharing.
Thanks for watching!
Best and stable strategy so far. thanks Ali.
glad you found it useful
This is great 👍 I have been looking for a robust way to trade the S&P instead of “buy & hold” - I’m glad I found your channel!
Hi Oasis! I wonder is there any efficient way to algorithmically trade stocks inside and of Stocks and Share ISA for me here in the UK, none of the brokers offer sophisticated order management is my problem
Glad it was helpful!
I think tradestation is available in UK, if not then you need to run an algo platform and enter orders manually in interactive broker, there are also services that take manual orders and send it to IB
Could this strategy be used on individual stocks going long-short?
of course, but only testing will tell the truth. next version of SQX will be able to test single strategy on multiple symbols, so that will be a good start on stocks.
Since the market always goes up (due to continuous 401k inflows), it might be advantageous to skew the long positions more than the short positions.
correct, also indexes are systems that continuously add outperformers and exclude under performers, that's why they keep drifting up
Ali, when trade a set of the underlying stocks instead of the ETF, did you divide the amount of capital you commit to each trade by the amount of stocks used in the portfolio?
So if you capital at any given time is 10000 and you are applying this strategy to 10 stocks, you buy shared for 1000?
yes
I really wish you would've showed a chart. All of don't have algo systems so we need to work the system manually but how can we do that if we can't see how everything is suppose to look? Can you keep that in mind next time?
Thanks for the suggestions
Code is in the community if you want to see how it works go.statoasis.com/community
Is it only making 1 trade until it hits TP?
Can this work on major Forex pairs?
Yes of course, but you need to pair it with a mean reverting currency pair
Hey, if you would pick 1 index strategy to use for investment in the next 20 years, which one it would be? Thanks
you should not trade one strategy, there are many studies that shows the power of uncorrelated strategies in a portfolio, even 3 strategies will make a huge difference www.statoasis.com/post/the-holy-grail-by-ray-dalio
There are strategies that are built like a portfolio, where they trade multiple assets using multiple factors, like TAA (Tactical Asset Allocations). those can be traded alone, there are many flavors of them if you search the term TAA.
I currently have 8 of them in the VIP+ channel in my community go.statoasis.com/community
Could this be optimized for lower timeframes? strategyquant is a little to expensive for me.
Also, how are you only getting an average trade of around $280, each contract is $25 a point and you are going in 100% with a 100k account. That around 180 contracts (assuming broker). So either its trading micro points or only trading 1 contract at a time.
test is on SPY (SP500 ETF)
smaller timeframe have a lot of noise
@@StatOasis oh, when i heard SP500, i was thinking ES
Excellent idea!
1.Is this strategy only long position?
2.A quantitative strategy that has been extended to a basket of stocks, a good idea. I hope to share more stock strategies developed by SQX in order to increase diversification. If this strategy can be extended to the advanced strategy of searching for the most suitable stocks for thousands of stocks, and only using SQX, I would be very interested in this.
3.Regarding a portfolio of 12 stocks, should I use $10,000 or 120,000 as a starting capital? Obviously these two values are unreasonable, so how much should be more reasonable? I already understand the starting capital selection method of a single strategy, but in the portfolio? Is it also through Monte Carlo simulation to check the ruin probability?
Capitalization can only be figured out properly through Monte Carlo testing on the portfolio
@@StatOasis Thanks very much， i watched the video "What is the best Money Management method for a portfolio?" again, and I have a new question and leave a message under that video
Yes it would have been interesting to see it trade long and short, say on futures. I'm testing it now.
For the stock strategy, did the strategy make another entry on each x day low, or was it one trade at a time.
Its one entry at a time, but you can add positions with multiple entries
How did you get data back to 1993?
Spy started in 1993
@@StatOasis ok, I guess I should've asked: To get the historical data back to 1993, are you subscribed to the SQ data package or did you import it yourself from a different source?
I have SQX data, and Tradestation data, both go back to 1993, SQX intraday go back to 2009, but I don't remember which one is used in the video.
FYI, if you want to subscribe to SQX data, use "StatOasis" coupon to get 23% off
If I understand the description correctly, you buy the dip on the 7th down day then sell when the asset hits a 7 day high, so it’s BTFD. How then are returns flat through all of 2008 and half of 2009? All BTFD strategies get slaughtered in bear markets because you’re continually buying back in on the way down
It looks like u didnt watch the whole video 😊
Adding a filter could easily mitigate that
where is the code?
I can not replicate base strategy results on spy. I am bringing Tradestation data into sqx. I can go to 60 k max while you go upto 82 k. What data source are you using? I have triple checked my settings to match yours.
this one was using SQX data provider SPY day session. not matching my results is ok, but you should be very close, as even same symbol from two data providers will have different results but very close.
check start/end date, variable shifts, daily session, etc.
super wow
Thank you so much
Are you on discoed still. I get re-directed to heartbeat.
Heartbeat only
Https://go.statoasis.com/community
Why dont you go with the connors RSI then?? Got 3 Indis in one,: RSI, number of down/up bars and distance to MA. Thank me later in Insomnia ;-)
CRSI is really good, but not available in SQX
@@StatOasis
What is CRsi?
@@iameladlevi Connors RSI
Everything was going well until survivorship bias on those individual stocks turned this into a clickbait.
Returns are actually a little bit higher if you include all delisted stocks.
I chose this route so viewers can replicate the test, but to include more expensive data that include all history for all constituents and platform that can do the test will be hard to replicate for my audience
@@StatOasis thanks for the response. is this strategy better than the rsi(2) on individual stocks?
It’s a different profile, but correlated to RSI2. Which one is better will depend on what are trying to achieve in your portfolio.
If you are looking for best returns on single instrument then it’s a lot easier to compare
I believe Norgate produces data that fixes the survivorship bias issue, possibly need to use it with amibroker as it`s integration with multicharts etc seems limited,,,would be interesting
With tradestation engine i get 13% DD and with MT engine 25%. Interesting ⁉🦙
U need different data for MT as they stamp the open while TS stamp the close