Timeframe for Automated Trading

How much time chart is suitable for automated trading?

I would recommend going MTF. With entries on tick charts. Greatly improves trading if used correctly.

SF

I've been having great success with the daily time frame. Every forex strategy that I've seen so far use the 5 minute time frame but I find I get a lot of fake moves using this time frame.

My own automated strategies use indicator inputs based on the M5, M15 and H1 timeframes (combined with pure price action). Sometimes I use multiple TFs in the same strategy, so I always make the timeframe an input parameter. This also prevents human error in platforms like MetaTrader when running the EA on the wrong chart: the program can now run on any chart. However, that doesn't mean these strategies trade at the bar open of these timeframes. Instead, each incoming tick, the strategy updates the indicator values and combined with price action decides whether to enter or exit.

Some commercial scalpers use M1 indicator inputs.

In the machine learning community things are different. There seems to be a preference to trade at bar open. I have seen algorithms for D1, H4 and H1 bars. Below that, the signal/noise ratio becomes too low to successfully data mine using the computational power available to the average retail trader. Big institutions can go to lower time frames, but estimating the data mining bias becomes really expensive at that level.

So good strategies can be found on virtually any time frame. However, as a beginner, start at the higher time frames as the signal/noise ratio is better there.

I don't think there is an empirical answer to your question, It will depend on your strategy, I think the previous tip from Yoriz was a good guideline, starting for higher time frames then if you like heavy metal start going down.

Hi mate ! Great tips above ! But don't believe anything we say, better do the work and find out for yourself, using what you find here as a lantern to explore new horizons, never to convince yourself of anything. Finding out by yourself and staying open-minded will not only help you find and build the necessary tools in your own path, but also give the confidence to know that it really does work and has worked year after year, and to responsibly move forward with it.

I have been in algorithmic trading for a while. Use mt5, put all TF as inputs, and experiment. You will get many surprises. Most of my strategies use at minimum 4 timeframes.... and use dynamic data, nothing fixed.

First, focus on one pair, one direction. Take a lot of notes. As Yoriz said, lower timeframes are impossible without great computational power. You need to analyze a great amount of data throughout a great amount of time. And I concur, Open price can and is used successfully, not only tick, but each strategy has its optimized version, and each need not be too correlated with another, etc. But be careful with open price. If using open price in strategy tester, your ea must also use open price when working live...

So many ways to skin a cat.... Be adventurous. Put even more time aside for it. Be careful with larger timeframes. Sometimes it seems like they are more reliable, but it is only because not enough occurrences are analyzed, and therefore comes with some extra risk, seldom realized before it's too late.

There is a lot more about this, but it's easier to discover as you go, especially since the strategies and approaches one starts with has a very high impact on the rest of thee adventure !

Good luck on your adventure mate ! Hope it helps !

Well, I think it depends on your own trading style! In my live trading, I am always suitable with higher time frames like H4, D1 etc! But I know, many traders are comfortable with lower time frames!

its kind of a loaded question. a better question would be what style do you trade?

are you intraday? a scalper ( a form of intraday) ? position? swing?

then that will tell you what time frames to use and yes as they said above me its better to use 2 time frames on an algo. the main time frame that tells the algo in which direction to enter on the lower time frame. i notice i get better results like that. I'm working on something like that now that's proprietary

I personally prefer the 15 min time frame as a day trader focusing more on multiple stocks, and the longer the watchlist is, the higher the cahrt interval should be.

Then how do you identify the market trend? I think, small time frames are too volatile to mark the market trend! I personally prefer the D1 for trend identifying!

15 minutes is good for trading, but all timeframes must be considered to determine price action. Trade with the trend but consider contrarian trade at market limits.
A watching brief is necessary to avoid overtrading. Indicators are not helpful except for stochastics.

I trade intra-day and usually do M5/M15 or M15/M30 (but i usually pick M5/M15 so that I have more actions in a day).
I follow those Price Action style trading to try to keep my chart clean with the with important indicators such as MACD/RSI/supp resistance zone.
The higher timeframe (M15) I usually use it to identify the zones and then I will just trade in the lower timeframe (M5).

If M5/M15 is too noisy, I will switch to M15/M30.
I don't do M1 as it is too noisy to trade and I can't really scrap well

For me automated trading works for intra day usually 15 mins but as a beginner level trader I haven't gotten the courage to experiment with my trades I like to play safe but there is no such thing as playing safe as far as forex trading is concerned. I think that is why I haven't been able to convert my most trades into profits. Sadly

a trending timeframe is pretty subjective. using the D1 for trend identification is for swing trading mostly though. it would be assumed that someone that uses the D1 for trend direction would be entering on the H4 more than likely. h1 is too far apart.

but if thats your style then thats what you build the algo around