 | MoniGoMani - Testing | 253 Members |
| Official MoniGoMani Testing Room where you can share your test results / config files / ideas / signal reports | 3 Servers |
| MoniGoMani - Testing | 253 Members |
| Official MoniGoMani Testing Room where you can share your test results / config files / ideas / signal reports | 3 Servers |
| Sender | Message | Time |
|---|---|---|
| 18 Feb 2023 | ||
 rikj000 | * Hmmm, idk if I would do it like that 🤔 That would add a 3th layer to the software stack (Freqtrade + Strategy/MGM + REST-Script) But the Strategy/MGM & REST-Script would not be able to share data/communicate with each other 🤔 | 16:04:14 |
| rikj000 | It would make implementation easier though 😄 Since then you're not limited to Freqtrade's strategy abilities, and have more freedom in your code | 16:05:27 |
| rikj000 | Perhaps it can be added in MGM-Hurry, which already runs as a 3th layer 🤔 | 16:05:53 |
 Meyo#1398 | It would be the Micro Services Architecture approach, as you don't have to worry about Freqtrades upgrades or whatever, it's standalone | 16:06:35 |
| Meyo#1398 | and does not need any information back, as it would just act as messenger.... | 16:07:14 |
| Meyo#1398 | so decoupled | 16:07:53 |
| rikj000 | It would be a nice addition to a Freqtrade setup either way of how it's implemented 🙂 | 16:08:06 |
| Meyo#1398 | yep, it would be something to consider | 16:08:28 |
| Meyo#1398 | Something like this perhaps.... '''import requests import datetime import time from bs4 import BeautifulSoup def is_meeting_in_progress(): """ Returns True if the Federal Reserve is currently holding a meeting, otherwise False. """ url = "https://www.federalreserve.gov/newsevents/pressreleases/monetary20220624a.htm" response = requests.get(url) soup = BeautifulSoup(response.text, "html.parser") event_date = soup.find("div", {"class": "col-xs-12 col-sm-8 col-md-9 col-lg-10 textleft"}).text.strip() event_date = datetime.datetime.strptime(event_date, "%A, %B %d, %Y").date() today = datetime.date.today() return event_date == today def main(): while True: if is_meeting_in_progress(): print("stop") else: print("start") time.sleep(60) # Wait for 60 seconds before checking again if name == "main": main() ''' | 16:33:41 |
| Meyo#1398 | * Something like this perhaps.... import requests import datetime import time from bs4 import BeautifulSoup def is_meeting_in_progress(): """ Returns True if the Federal Reserve is currently holding a meeting, otherwise False. """ url = "https://www.federalreserve.gov/newsevents/pressreleases/monetary20220624a.htm" response = requests.get(url) soup = BeautifulSoup(response.text, "html.parser") event_date = soup.find("div", {"class": "col-xs-12 col-sm-8 col-md-9 col-lg-10 textleft"}).text.strip() event_date = datetime.datetime.strptime(event_date, "%A, %B %d, %Y").date() today = datetime.date.today() return event_date == today def main(): while True: if is_meeting_in_progress(): print("stop") else: print("start") time.sleep(60) # Wait for 60 seconds before checking again if name == "main": main() | 16:34:02 |
| Meyo#1398 | or maybe it can be added directly into the strategy script.... | 16:38:11 |
| Meyo#1398 | but if there is an error, website changes its design or whatever, the strategy would crash, so maybe keep it separate. | 16:38:55 |
| rikj000 | Something like that yes 🙂 In case of MGM it would make sense to completely stop the bot and notify the user on meeting start/ends (through telegram/email/...) So that they can re-optimize the framework depending on the outcome of the meeting (bullish/bearish scenario) | 21:42:25 |
| rikj000 | * Something like that yes 🙂 In case of MGM it would make sense to completely stop the bot and notify the user on meeting start/ends (through telegram/email/...) So that they can re-optimize the framework depending on the outcome of the meeting (bullish/bearish scenario) | 21:42:27 |
| 19 Feb 2023 | ||
| Meyo#1398 | yep, and have a weighting for that, so it may avoid placing buys if the meeting is bearish, oh, and another thing to consider, I think the hours before the meeting is always bearish, as we are not the only ones pulling the plug on bots.. so better stop the bot well in advance.. | 08:55:24 |
| Meyo#1398 | unless its shorting | 08:55:26 |
| rikj000 | Doubt them to be always bearish, really depends on the macro economic situation at that point in time 😮 But depending on the outcome of the meetings, the market tends to shift upwards/downwards for a while. So imo it's okay to "miss out" on the potential profits of an initial pump/drop right after the meetings, since it will be in trade for more certainty about what the upcoming days/weeks will bring. Which then gives us time to optimize our bots/frameworks accordingly! 😄 I do agree that stopping the bot some time before the meeting starts is a good idea though 🙂 Since during the meeting can be volatile/dangerous. The meetings usually last 1 - 3 days, so I would stop the bot maximum 1 full day on beforehand, but probably less, guess a few hours on beforehand will already be enough. | 10:21:40 |
| 1 Mar 2023 | ||
 Cynthia D. Tolentino joined the room. | 20:28:44 | |
| 3 Mar 2023 | ||
 rUpere joined the room. | 09:12:10 | |
| 7 Mar 2023 | ||
 s.d.d. joined the room. | 14:58:14 | |
| 11 Mar 2023 | ||
 rUpere | Rikj000, Meyo Assuming FED meetings happened in the past and we know the timeframes - theoretically it is possible to analyze this scenario you are both describing. Backtest, then plot the trades + the FED meeting for visual interpretation. Export entries/exits that were made during(+/- few hours) the FED meetings. Then analyze how those specific trades are really impacting the profits. Next step: implement in MGM backtesting a mechanism that will create an exception during past FED meetings to see if profits are improving. Sure more analysis can be done if there is enough data. But in this case all necessary data is available to make a better decision about what you are describing above. Let me know about your thoughts. 😉 | 18:05:03 |
| 13 Mar 2023 | ||
| rikj000 | The goal of following the FED meetings is to figure out ahead of time if they will raise interest rates (kills stocks, pumps dollar), or if they will decrease interest rates (pumps stocks, kills dollar). The volatility of the meetings is a side effect of the market trying to predict above. MGM wouldn't yield much benefit out of analyzing said volatility imo. However if we could get signals, based on sentiment analysis of the FED speaker (e.g. saying disinflation 15 times during the interview after the meeting = bullish), that would be very use full. However I think it will be very hard to program such a feature. | 04:34:23 |
| rikj000 | * The goal of following the FED meetings is to figure out ahead of time if they will raise interest rates (kills stocks, pumps dollar), or if they will decrease interest rates (pumps stocks, kills dollar). The volatility during the meetings is a side effect of the market trying to predict above. MGM wouldn't yield much benefit out of analyzing said volatility imo. However if we could get signals, based on sentiment analysis of the FED speaker (e.g. saying disinflation 15 times during the interview after the meeting = bullish), that would be very use full. However I think it will be very hard to program such a feature. | 04:34:55 |
| rikj000 | * The goal of following the FED meetings is to figure out ahead of time if they will raise interest rates (kills stocks, pumps dollar), or if they will decrease interest rates (pumps stocks, kills dollar). The volatility during the meetings is a side effect of the market trying to predict above. MGM wouldn't yield much benefit out of analyzing said volatility imo. However if we could get signals, based on sentiment analysis of the FED speaker (e.g. saying disinflation 15 times during the interview after the meeting = bullish), that could be very use full, if implemented correctly. However I think it will be very hard to program such a feature. | 04:35:39 |
| rikj000 | * The goal of following the FED meetings is to figure out ahead of time if they will raise interest rates (kills stocks, pumps dollar), or if they will decrease interest rates (pumps stocks, kills dollar). The volatility during the meetings is a side effect of the market trying to predict above. MGM wouldn't yield much benefit out of analyzing said volatility imo. However if we could get signals, based on sentiment analysis of the FED speaker (e.g. saying disinflation 15 times during the interview after the meeting = bullish), that could be very use full, if implemented correctly. However I think it will be very hard to program such a feature.. | 04:35:49 |
| rikj000 | It will require to hook up MGM to a sentiment analysis AI, which has the ability to scan a video/audio feed for bullish/bearish news, and then notify the MGM framework somehow about the results. | 04:37:01 |
| rikj000 | * I guess it will require to hook up MGM to a sentiment analysis AI, which has the ability to scan a video/audio feed for bullish/bearish news, and then notify the MGM framework somehow about the results. | 04:37:26 |
| 16 Mar 2023 | ||
 Stas joined the room. | 09:20:22 | |
| Stas changed their display name from Stanislav Prokopenko to Stas. | 09:30:07 | |
 Stas joined the room. | 11:02:05 | |