Support Board
Date/Time: Wed, 27 Nov 2024 18:42:23 +0000
[Programming Help] - Looking for an Algo Engineer
View Count: 386
| [2023-07-07 19:37:39] |
| User373245 - Posts: 42 |
|
Looking to hire someone who can code in Sierra charts ACSIL, C++ and meta trader 4/5. We are trying to convert our simple alert into a fully automated trade in meta trader and have some requirements that need to be engineered on a more advanced level. First. We need the ability to build the alert on Sierra charts to a buy or a sell and dump a number of studies data reference into a spreadsheet relative to that alert. We don’t need help building the alert…. just need the ability to both build and adjust it as we see fit. So, say I have a 7 cross 21 moving average alert on a 240 minute bar and it triggers at 11:33:22 and I would buy. I want to know what the number of traders are at on a 30 60 240 min chart at that exact time as well as say what the standard deviation value is in a 240 min chart. So we need the ability to associate a cell column in the spreadsheet relative to each study as well as ad studies as we sit fit. But then each cell must be able to dump data at the exact time the alert condition is met into the spreadsheet and not the end of bar. Then we need to know the high or low depending on the buy or sell. Or apex relative to direction from trade entry to the end of day. If it's a sell, the low of the day. If it's a buy, the high of the day in pips but only the pip value relative to after the alert triggers, to end of day. Next we need the drawdown in pips from the price entry relative to a stop loss and limit. So say we have a 35 pip stop and a 40 pip limit and on a sell trade and the market rolls against the direction of your trade expectancy, how much in pips does it drawdown against you before it hits the limit. The spreadsheet should have all the relative data of trade data. Asset name, Price entry, Price exit, Time, date, direction (buy or sell) Apex, Drawdown, and all study refernces. Pivot value % and lable Next we meed the ability to push the simple alert to a buy or sell to a mt4 (it doesn’t have to be an EA just the ability to connect the alert generated from Sierra charts to the broker and have a trade precedence. We are aware of the broker API already built in SC but use brokers outside of this spectrum. This is where it gets a little complicated as the exit is based on daily pivot points and have a couple caveats which can be defined relative to a pivots %, label and a couple if statements. IF the pivot is less than 18 pips in value from one to another, two pivots. But the exit is not the exact value relative to the pivot label. Pivot exits as a probability are higher at call it 90% of the pivot. The ability to adjust this value would be good. As it needs a bit of testing relative to engineering execution and we need to be able to run some data to get the exact. SO the ability to adjust it as well as dump it into the spreadsheet. So say you have an alert condition trigger a buy and its 20% past pivot r1. We need to dump the label r1 and what % past the pivot its triggered. Call the distance of one pivot between the other 100. As far as the exit to the trade is concerned we will have to define a set of parameters relative to where the alert triggers to the pivot %. IF within lets call it 20% exit that pivot unless the value is less then 18 pip then 90% of the next pivot label. The if value to less than 18 will be universal every trade. If the entry is within let’s call it 45% and 55%(the middle of the pivot then exit is the next pivot at 50% or the middle. If the remainder of greater than 55% is equal to 18 or greater to 90% of the pivot value we can exit there. If not we should exit at the same value % on the next pivot. SO say we trigger at 58% from r1 and the distance to 90% of r.2 is less than 18 pips then we would exist at 90% of > r2. The ability to adjust this from call it 18 to 20 would be nice. As it will depend on the election of the automation relate to the math that is the exact value but we can start at 18. Like where it triggers relative to the pivot, relative to what pivot I exit. As when we trade live if the alert triggers at the beginning of the pivot, call it R1, we take the trade to about 95% of the next pivot, in this case to R2. So essentially I am trading a single pivot per trade but it depends on where the alert triggers from relative to the pivot. If the alert triggers at the end of the pivot I will go into the next. So if it triggers at the end of call it S1, I exit about 95% to R2. If it triggers somewhere in the middle, I exit in the middle of the next pivot. This will have to probably be done relative to a % value of the pivot on the exact nature of how this will be coded. I am not 100% sure what this value % is mathematically, as it's a feel relative to the visual nature of looking at the pivot, so maybe we can dump this into the spreadsheet as data relative to the % and label of the pivot. So alert triggers 20% of S2 or 72% of R3 and runs to X % pivot label? All of the trades are day trades but we have a weekly alert coded on a weekly chart so that if the alert happens or within the trade duration of the day trade and the weekly alert comes in within the trade duration or it is present at time of entry, we let the trade run the full length of the day and liquidate 1pm EST but after the trade has run to 40pips move the stop to break even. Then the weekly criteria is null for the remainder of the week. So if this fundamental occurred on say Tuesday from a sell EURJPY and another sell comes in on the same asset on Thursday, the weekly criteria is null for the rest of the week. It only works once a week and the first time it occurs. But also if the weekly triggers on say Monday and I don’t get a day trade alert on the day it triggers and then get a day trade alert on Tuesday, I don’t take a big trade. The weekly alert is never an entry. It is simply an indication of high volatility and to take a bigger trade but they are separate mathematic fundamentals and must occur at the same time to work. So essentially if a weekly alert triggers it must null as a big trade every day after for the rest of the week. Regardless of weather it times with a day trade. But if it does time with a day trade we need to be able to take a big trade from it. If my weekly has an alert in the opposite direction, I need to null the day trade alert and not trade it. I use the weekly to decide if I take a big trade, not to enter. But I would never trade against it but only relative to the alert I have built on it (maybe there is a way to code this form the SC side). So say I have a day sell EURAUD and the weekly has a BUY color bar, null the day trade alert. I would never trade against the weekly if its at a certain value going against the direction of my trade. I already have this alert condition built out. Most trades stop are 35 stop some GDP pairs have a 38 stop but this is also why I would like to study drawdown in a spreadsheet, so I can find the average drawdown of years of trades relative to each asset and find a more mathematical number here. So the ability to replay years of data relative to the alert is necessary from the automation. I don’t need to replay to mt4 I can do it from SC replay and get data that way. All trades are a day trade and liquidate at 1pm EST if neither hit stop or limit. We never take trades past 11am EST. Then we need the ability to set a limit to how many trades we take per day, never two trades in the same direction on the same day on one assist but we can trade opposite side. So say you get a buy on EURJPY 9pm est and another buy come in on EURJPY at 3am…you can not trade it but your could if its a sell trade. Most likely the limit of trades in the market at a given time will be 4 but we will need to play with this some relative to the capital and risk metrics, so maybe a place to change how many trades at a given time. Then we should have something relative to the limit order relative to the price the alert triggers at and maybe a time duration for this. As we do market orders when trading live. Let’s call it 4 pips from the price the alert triggers at relative to the execution of the order with spread and market difference from our data feed to the broker. Id rather not do it as a limit order as I can give it a bit more rom then that exact price. When I do this live, it’s a matter of waking up and getting to the the computer and placing the trade. And 95% of the time I am ok at a market order. If the market is racing and too far from my alert price I just won’t take the trade. But also I should account for the broker being an A HOLE! A limit order is too tight! I’m ok with a little adjustment of 3 or 4 pips. I am not 100% sure on the time frame so maybe we can play with it some and adjust it? Maybe start with 3 minutes and 4 pips? But also if we do not get in a trade for this reason you still must treat it as a trade an null all alerts on that asset in that direction for the day. We need to be able to adjust our lot size relative to account growth which we can do manually. As you can see there is a bit of adjustability we need to be able to do. One of our traders has a bit of coding experience and so we can do it from the coding syntax but also the ACILS will give you a UI to do all this from too if that is an easier approach. This is a relative idea of what we are looking for and open to discussion to execution and some of it is a bit complicated so more then happy to talk a bit more and do a screen share for a better explanation. But you must be able to work with Sierra Charts and must be able to work with meta trader. |
| [2023-07-08 13:09:29] |
| Sawtooth - Posts: 4120 |
|
Choose one here: https://www.sierrachart.com/index.php?page=doc/SierraChartStudyAndSystemProgrammers.php Date Time Of Last Edit: 2023-07-08 13:10:20
|
To post a message in this thread, you need to log in with your Sierra Chart account: