MTF – Simple question – Please help

Forums › ProRealTime English forum › ProOrder support › MTF – Simple question – Please help

Viewing 13 posts - 1 through 13 (of 13 total)
  • 03/23/2021 at 9:46 PM #165103
    Khaled

    Hello everyone,

    May be very basic explanation, but I’m struggling to see clearly the origin of my problem.

    If I understand MTF correctly, you can build a code as follows and expect that each time the Close crosses the HullAverage[13] on 1 hour TF, an order gets executed on 1 min TF. Right? (just an example, not meant to be profitable or whatsoever)

    Unless, I missed something, the result doesn’t look to be okay.

    Short : If I look at the short order exectued on 23 march at 06:00 (1 min graph), the Close was below Hull Average on the 1 hour TF, but it was not a point of crossing close vs HullAverage on the 1 hour TF.

    Long: If I look at the first blue bar on 1 min chart (right hand), the Long Order is executed on 23 march at 06:08. When I check the main graph (1 hour, left), le close was below the HullAverage and didn’t cross over.

    If I looked at the crossing over close at 07:00 (1 hour TF), it doesn’t correspond to a Buy order on the 1 min TF (the system was already in long position since 06:18, where close on the 1 hour TF was under the Hull Average).

    Can you please help me understand how to make PRT buy and sell on 1 min TF when conditions are met on 1 hour?

    Thanks

    03/23/2021 at 10:12 PM #165113
    Khaled

    To clarify further: I don’t want to wait until the 1 hour TF bar closes (no UPDATEONLOSE). As soon as the condition is met during the hour, the order should be executed on the 1 min TF.

    03/23/2021 at 11:01 PM #165119
    Khaled

    Apparently, if the Close crosses the HullAverage in 1 hour TF multiple times and in both directions, it created multiple Long and Short orders during the course of the hour. So, when we look at the candle already closed, we don’t necessarily realize that crossed both sides.

    Is there another possible explantation please? any simple way to filter this?

    Thanks

    03/23/2021 at 11:35 PM #165121
    Vonasi

    You have chosen DEFAULT for your slower time frame so there are 60 one minute candles in every hour where your conditions could be met. CLOSE can cross over or under an average once every minute – multiple times!

    Add the following to the bottom of your strategy to fully understand when each condition is true:

     

     

     

    1 user thanked author for this post.
    avatar Khaled
    03/24/2021 at 12:20 AM #165125
    Khaled

    Thank you Vonasi.

    Should be much simpler then to put all conditions and execution of orders under TIMEFRAME(1 hour, DEFAULT). So, once a condition is met and a trade is executed, crossings on lower TF don’t get me out?

    On backtst, the results are much better on higher TF.

    03/24/2021 at 11:19 AM #165146
    GraHal
    Khaled wrote:

    put all conditions and execution of orders under TIMEFRAME(1 hour, DEFAULT).

    If you do above then you could not have a lower timeframe (no code to exist under lower timeframe??) and so the code would be read once per hour only.

    So results for above would be same as running without using MTF, and running on a 1 hour timeframe.

    1 user thanked author for this post.
    avatar Khaled
    03/24/2021 at 1:34 PM #165169
    Khaled

    Thank you GraHal and Vonasi for taking the time. Another simple, slightly unrelated question.

    If I want ProOrder to chose between average[n] and n are Fib numbers, i.e. 3, 5, 8, 13, 21, etc. To reduce the backtest time i don’t want the syzem to test all values betwwen 3 and 21. I would also like to avoid to code like if close>Average[3] then, elsif close>Average[5] then, etc…

    Is there a simple way to ask the system to chose a predefind list of variables?

    Thanks

    03/24/2021 at 2:15 PM #165175
    GraHal

    Yes … use the Optimiser

    https://www.prorealcode.com/blog/video-tutorials/how-to-optimize-a-trading-system-with-probacktest-prorealtime/

    Put in values as min 1 max 21 step 1 and disregard / don’t use  values not in your fib value list if they come up as profitable results.

    Or maybe, better still, dont disregard non-fib values as it may mean you get out before the trade hits a fib level (when everybody bails out) then you may become a millionaire before the rest of us! 🙂

    03/24/2021 at 5:18 PM #165201
    Khaled

    Thank you GraHal, becoming millionnaire from Trading is not yet on my shopping list. I’ll ask Santa at next Christmas 🙂

    My take away after a few months of trail and errors:

    1. day trading, manual or automated, is a full time job. Buy and Hold a few indexes S&P500, NQ, DJIA, DAX is more profitable on the long run. Making 1% per day (0.5% with 2x leverage) over 150 days per year would make you 4.45x of your capital, but still 76% of CFD traders on IG are losing money… where is the mistake?
    2. any set of “standard” indicators” with standard settings can get you good entry points… you can play around the TFs to further optimize.
    3. it’s more difficult to determine when to get out of a winning or a losing trade than to take the trade itself.
    4. money management and psychology are the two most important variables that one needs to master  before diving into the piranha swimming pool (Elder mentions the piranha in one of his books)
    5. if automatic trading was that simple, banks would not hire traders for millions – (Google, Facebook and the like work on very sophisticated AI, but I didn’t hear about any of them getting into automated trading)
    1 user thanked author for this post.
    avatar GraHal
    03/25/2021 at 12:14 AM #165247
    Vonasi
    Khaled wrote:

    if automatic trading was that simple, banks would not hire traders for millions – (Google, Facebook and the like work on very sophisticated AI, but I didn’t hear about any of them getting into automated trading)

    Banks don’t hire traders for millions. They get very good university students to work for free with the lure of a potential large salary and give them a tiny amount of someone else’s money to play with and see how they do. A tiny few will go on to earn a small percentage of the millions profit that they make risking other people’s money – which might just add up to millions if they are clever or lucky.

    For us minions we are just betting with our own money. Will price go up or down from here and how far up or down do I think it will go? It is the same as asking ourselves which horse will win a race and how much will it win by. When we bet on a horse we read the form book and see what happened in the past and make our decisions based on this. When we spread bet we look at indicators and see what happened in the past. The bookie/broker is doing the same thing but they have the edge because they decide the spread/odds for each horse. In spread betting our first job is to beat the spread and in horse race betting it is to look for an unbalanced book of odds that favours our horse more than the others.

    At the end of the day we have to keep in mind that we are betting… not investing. Banks are betting too but using some clever people they found with our money and not theirs!

    1 user thanked author for this post.
    avatar Khaled
    03/25/2021 at 12:45 AM #165249
    Monochrome

    I think banks are using much more sophistcated trading tools not available to us mere mortals.

    Automated trading as a way of replicating live day trading is hard work. Sometimes intuition is hard to code for.

    Ive been on this for a few months too, lots of trials, all failures. Its a nice learning experiance and will continue this as a hobby incase I strike gold with something in the future xD

    1 user thanked author for this post.
    avatar Khaled
    03/25/2021 at 9:07 AM #165274
    Khaled

    I’m not claiming to be an expert… Just a humble opinion of a small, still unprotibale, wannbie day trader…

    May be I don’t have the full picture, to my knowledge, when it comes to autmoated trading, Banks and Trading Instituions use High Frequency Trading tools, where the speed of the internet link is as important as the Algo themselves. They “bet” large amounts and expect only a small variation of 0.00x% within seconds…

    What I meant in my previous post, Banks would n’t hire costly traders, even if it’s only bonus based reward, it’s still cash out…

    Have a good day!

     

    03/25/2021 at 9:15 AM #165275
    Khaled

    In your experience, what’s the best Money Management code for day trading? (please post a link)

    I’ve tried many posted by other generous contributors… overall, each code is more suitable for a certain type of situation… i.e. some codes are better suited for Trending market, others are better suited for scalping (or small steps), other better suited for range…

  • Author
    Posts
Viewing 13 posts - 1 through 13 (of 13 total)

Find exclusive trading pro-tools on ProRealCode Marketplace

Similar topics:
rising moving averages – please please help

Create your free account now and post your request to benefit from the help of the community
Register or Login