ProRealCode MarketPlace: sell your trading products to thousands of ProRealTime users!
Forums › ProRealTime English forum › General trading discussions › ProRealCode MarketPlace: sell your trading products to thousands of ProRealTime users!
- This topic has 551 replies, 88 voices, and was last updated 4 months ago by
Mauro T. “Algorithm System”.
-
-
12/01/2021 at 4:21 PM #182625
without buying a new license
So I’d buy a new license, no big deal or difficult.
The inability to remove ‘expired license Algos’ from my Platform has caused me to take a step back and rethink.
I find having a clear list for Algo backtest makes me so much more ready to install an Algo for which somebody is asking for help etc.
In the past, when I had dozens of Algos listed I used to think … oh No, I don’t want to add another to my List, I can’t find what I want now out of my own Algo List.
Also, I no longer have Issues with platform freezing / hanging; such Issues ceased (for me) when my Algos listed are kept < 12 or so. Mostly I have zero / nil Algos listed as I store them all on my PC.
Now, if I trial MarketPlace Algos, I would be forced into having Algos listed on my Platform for the next 20 years and would repeatedly curse them for overloading my backtest list etc.
I am having a rant as I feel strongly about this.
I hope you can think of an ‘easy way’ (for you) round this Nicolas?
Thank You in Anticipation
GraHal1 user thanked author for this post.
12/01/2021 at 4:29 PM #18262612/01/2021 at 4:44 PM #182629Thanks Nicolas for the answer, we saw it today.
We believe that it is very important for users to know the total number of updates to a product, and the dates of the updates.
Otherwise, it is very easy to cover up detrimental operations by algorithm providers and misleading advertising about the alleged benefits.
In this way we protect users from possible fraud.
Thanks!
1 user thanked author for this post.
12/02/2021 at 8:58 AM #182661Although I understand perfectly what you want to express, I rather think that the availability of the changelog proves the follow-up and the efficiency of a vendor. According to you, the update of an algorithm should not change the backtest of the past that took place before this update? The simple modification of the period of the moving average of the indicator that establishes your trend will have this impact, the seller will modify it simply because the market regime has changed and he has to adapt it, it is his duty as a supplier to his customers, and in this case we can’t talk about “fraud”, right?
To solve all these questions and debates, I have in mind to create a real time tracking of strategies on a third party site, a kind of labeling to allow anyone to follow the trading history, no brain filter here, just some months of work 😉
12/02/2021 at 9:23 AM #182665Wow that sounds a fantastic idea Nicolas, take care and don’t make yourself ill (again?) over all this extra work!
When I dive into MarketPlace, after a few minutes my head is spinning with all the different ways Vendors promote their products. This makes it hard to compare one Algo with another etc.
The good thing about tracking on a third party site is that performance charts / information would be standardised across all Algo’s.
1 user thanked author for this post.
12/02/2021 at 9:27 AM #182667The good thing about tracking on a third party site is that performance charts / information would be standardised across all Algo’s.
Exactly, and I’m eager to complete this work, but still a lot to do!
5 users thanked author for this post.
12/02/2021 at 9:59 AM #182671Good morning Mr. Nicolas, and friends!
Indeed, you are partially right, it is true that market conditions can change drastically and require updating algorithms.
In fact, privately, from Skynet we let you know a few days ago that at least one of our bots will probably be updated soon.
However, we will let all foreros know the changes made. We have to understand that updates can be a double-edged sword:
-Updates can be code improvements.
-The updates can be concealment of past operations with damage, and in this way keep the backtests “clean” for future users.
To our team it seems a very good idea that users can see the operations of the bots !!!
Do not stress, Christmas is coming, it is time to enjoy, and with family and friends.
😉
1 user thanked author for this post.
12/02/2021 at 10:47 AM #182687-The updates can be concealment of past operations with damage, and in this way keep the backtests “clean” for future users.
Excellent initiative Nicolas.
Regarding the past Backtests I don’t think there are people who look at the backtests to buy a product so even if you improve them what they are used for ….. instead as Nicolas says there will be a moment that you will have to tweak or improve an algorithm.
I believe that those who buy want to check the operations that have taken place for months to have confidence in what they buy …. then like all algorithms they will have excellent moments and negative periods, but this is part of trading.
For the Backtest it is sufficient to go to Prorealcode there are 197 strategies all Backtest and free.
Maybe it’s just my thought and there will be someone watching Backtest to buy.Hello
Mauro T.1 user thanked author for this post.
12/02/2021 at 11:44 AM #182701Mr. Mauro.
Regardless of what has been written so far on this subject, in the end, among other things, we suppose that what is best is to demonstrate the proper functioning of the algorithms with real benefits.
Something that we are trying to demonstrate on Skynet, and you in particular.
Although, we also believe that users do rely in part on the backtest that has passed.
You would rent a bot that in the backtest has passed shows a loss in the account of € 10,000 ?, or that sometimes has drawdowns of 70% (examples) … we suppose not.
Finally, at Skynet we believe that users should have the right to know the dates of the updates, and the total updates.
In addition, and if possible, the reasons for the updates.
12/02/2021 at 11:57 AM #182704Notice Mr. Mauro that sometimes some foreros have asked us for a backtest of 1M units, thinking that if in such a long term the account is positive the bot is reliable.
The users see, and bases, the backtest to decide for buy a bot.
Although algorithm providers know perfectly well that such a long-term backtest is useless.
Notice the difference between the years 2018-2020 backwards and forwards.
Those years have been a before and after in all strategies in the stock markets.
12/02/2021 at 12:45 PM #182709If as mentioned before you look at the 197 strategies in Prorealcode at the publication date of each strategy you think you will find a backtest with losses or worse with a 70% drawdown … they will all be perfect where one says in three months I will buy a new machine and then I buy the new apartment ….. they also ask me for backtests and rightly so I am in the store, but as I said before and I repeat for ME the operations performed and update all are only important, both when the algorithms they are good both when they go wrong ….. I would say to be honest towards possible buyers, and when I re-optimize an algorithm I put the previous and current results not the previous and current backtests but the operations performed …… then in life there are people who commit suicide and I would say that it is certainly less painful to commit suicide in trading.
But I repeat this and only my way of seeing trading, then we are all different and if it were not so how sad life would be.Hello
Mauro T.12/02/2021 at 3:27 PM #182726Correct Mr. Mauro.
You and we try to ensure that users have sufficient information about the algorithms and the benefits.
But think about it… If we see an algorithm that is updated, let’s say monthly, it gives a lot to think about, and if in addition, the providers give little or no information about the updates … something smells bad there.
1 user thanked author for this post.
12/02/2021 at 5:13 PM #182738Some time ago here in the forum many complained because so many sellers did not put the LIVE, and Nicolas replied repeatedly that it was the possible buyer who had to ask everything that interested him most from the seller (in this case the LIVE) ….. now it is the same, if one re-optimizes every month or even every 15 days it will always be the possible buyer to ask the reason and be seen all the past if he does not do it it is because it suits him …….. or he wants to commit suicide .
Then as I said above we are not all the same, you cannot know if one buys precisely because it is always re-optimized every week and believes that it is the best way to always be updated on the market ……. Nicolas has already been clear about it .Hello
Mauro12/02/2021 at 5:20 PM #182740to close the debate, as I mentioned, the best way to keep track of the performance of the strategy since its launch in real and to certify this history of trades by a third site, I’m working on it, but I have no short-term solutions, apart from those I have already mentioned a thousand times as Mauro points out, the strategies on the marketplace are traded live, no seller can offer a strategy that is not. Before buying a product, there are several possibilities to contact a seller from the site, forms are available.
12/03/2021 at 8:58 PM #182793Good afternoon!
We at Skynet strongly agree with Grahal.
The buying and selling system is complicated for both suppliers and users.
Too many options, maybe it should be more standard.
On the other hand, and as we have already commented on other occasions, we believe that the one-month trial period is too short to demonstrate anything, or practically nothing, to users.
We believe that three months at least is a minimum period to start drawing conclusions.
Thank you and good weekend!
-
AuthorPosts
Find exclusive trading pro-tools on