Indicators working on different time scales

[Iotrez]

Hi all,

I was wondering if different indicators are better suited to different time scales.

For example moving averages....

When the current value curve crosses the moving average line it indicates a change in movement, but is this the case whether the time scale is both 1 day and 1 month?

Thanks.

 

[damianoakley]

Hi lotrez,

Whilst I am not a big user of indicators, to answer your question, I would say that most of the classic indicators like Moving Averages or Stochastics can be used as a tool across a wide range of timeframes.

There are many times when you can be looking at a chart, and if you took away the labels that showed time/date and price, then you probably wouldn't know whether you were looking at a 15min chart or a daily chart. That's not true of all charts of course, but is interesting none the less.

There are some specialised indicators which are probably better suited to shorter timeframes, and I'm sure that some of the day-traders on here can point some of those out to you.


Thanks

Damian

 

[Trader333]

It would make more sense to use certain indicators across several different timeframes rather than the other way around in my view.


Paul

 

[TWI]

If you are using a short term time frame to generate signals it can often improve results if you use something like a daily MA as a filter. It is not always the case but I have found it useful when using mean reversion strats but not breakouts.

 

[options]

And also keep in mind that one ma can be very similar to another ma. By that I mean
for example a 34 wma is very similar to a 22 ema on say a weekly chart.

And if you place a few ma's on one small time frame chart then move out to a longer time frame, one small number ma on the small frame will take the place of one of the bigger number ma's on the longer time frame chart.

(That'll keep you going for hours).

So how to pick the right ma?

Keep to one time frame and stick to it. Go back timewise as far out as you can and pick whatever ma you want, then custom fit it to the chart. Zoom back in bob's your aunty.

 

[grantx]

Lotrez,

For what they are worth, these are my thoughts.

For a simple example, assume you’re using 15 and 60-minute ma’s for possible buy and sell signals. Sometimes it works, sometimes it doesn’t. I think the reason false signals are given is because the changes on the underlying are not constant nor smooth. In other words, the ma’s have to be fine-tuned or optimised regularly to account for the changing volatility of the underlying.

TWI,

“…using mean reversion strats but not breakouts.”

Could you expand on this, please?

In the context of scalping, I’m currently looking at limits of price movements (and their standard deviation) over time intervals rather than actual price levels. Interesting results, yet to be formalised.

Grant.

 

[options]

I think what TWI is saying is that he uses reversion strats which are a rubber band approach.
IE: Once a price has broken out and moved too far away from a mean price average or a linear regression average he would then trade the pull back to the mean average of the price.

Roughly speaking it is when a price has shot off quickly and then exhausts itself. IT WILL move back to the normal price average.
It means that you miss the first move of the price but get the smoother pullback and quite often, (very often) the 2nd move in the same breakaway direction of the original will give you a better run for more profit.

The first run being a test of supply and demand from the people that matter. Who don't want the likes of me or you in the trade. It serves two purposes. Firstly, it wipes out the eager beaver newbies (on the pullback) allowing the main players to load up even more, before it then takes a quality move in that very first direction.

 

[grantx]

Options,

Thank you for the reply.

There are parallels with what I'm looking at. Where they may be some variation between my and Option's intentions is that I'm not looking for direction, but simply a repetition of degree of movement.

This is derived from observing that where a general trend is relatively smooth, ie lack of volatility on the underlying, the high-low range of bars tends to be (hopefully) reasobably constant. For example, if 15-minute bars are averaging 10 points, then one would look to buy (sell) at what may be the bottom (top) of a bar for a quick scalp. That's the theory, and limited experimentation has supported this. However, still plenty of fine-tuning and investigation to be done.

Tying this in with my remarks regarding re-optimisation of ma's in my first post, the time period for bars needs to treated likewise. For example, 15-min bars work one day, but 10-minute bars may be more approriate the next day, ie the underlying is more volatile.

What I haven't done yet, and I'm pretty certain there will be mileage in this, is to look at how options' implieds (skew, curve) unfolding throughout the day impacts, reflects or correlates with regards to the underlying volatility and or direction.

However, running all this, especially with Excel/dde's/VBA components simultaneously and in real-time, will really push my pc's capacity to its limits. We live in interesting times.

Totally irrelevant, what do the stars mean below one's name. Staff at Wimpy bars had these, presumably designating seniority. Always reminded me of the Order of Lenin.

Thanks, again Options. Hope the weather isn't too hot for you over there. I'm not bitter - it's in the low 70's here in the UK (I think there may be zero too much there).

Grant.

 

[options]

Okay, just to give a near real time example.

Cable , today. Market heading in a smooth down line. 1.30 uk time on the dot. Price bombs 1.9380 high of the ten min price bar down to 1.9352 or therabouts. 1.9346 was hit ten minutes later after the conductor had clipped all the late comers tickets. It then proceeded to rubber band back to 1.9375 almost 75% retrace of the down move. (It almost alway goes 50% at least).
Then preceded (After most are lying dead and wounded to fall to 1.9321All in the space of exactly 2 hours. (The last move being a further 25% more than the original first move.

The moral? Most times wait for the 2nd bus It is less crowded with sightseers and more comfortable.

Okay. How to help Grant.

"There are parallels with what I'm looking at. Where they may be some variation between my and Option's intentions is that I'm not looking for direction, but simply a repetition of degree of movement."

Trying not to hijack this thread, but you run the risk of becoming an analyst and not a trader.
At slow times there will appear to be a repittion of movement. (Until you come to try to trade it)

You must look for when the market is due to move. By this I mean time and price. Time is when a market will move. ie on open, and news. Those are the only 2 times you should be ready to trade. (Although it may be many times during the day news wise). At other times you are free to analysis to your hearts content).

If you can spot a pattern in a slow moving market it will without doubt come back to bite you in the bum. The open and the main news are market turning ops. Be ready to trade them.

Don't waste your time in trying to re-invent the wheel. Copy what works and take it from there.

I would point you in the direction of the Vegas 1 hour method and take it from there.

Every single trader going thinks he can get a different slant on things and get an edge that way.
Don't let me discourage you. You might be the one that 'cracks' it and more power to you. And despite what I say you will carry on anyway.

But it seems that the people that make money are the ones who have realised this and trade on price action in the majority of cases.

All you need is a price chart and a single straight line to win.

(Please do not think of me as being too harsh, but I have been doing this now for more than a few years and you really do not have that much time to waste).

The stars mean that I can make a really good burger, and the weather over here in Tenerife is fantastic. It is 5 in the evening and and still very nice sunshine, bit cold though I think it is around 19 c

Kind regards.

 

[grantx]

Options,

Over-analysis is a fair point. But as there are hundreds of indicators, each used differently by all traders with their own methods, it’s a question of identifying whichever is appropriate to my objectives.

And while over-analysis may seem somewhat laboured, the result may be only a couple of numbers albeit constantly changing. So there is little difference between two percentage numbers, for example and two ma numbers. It is more a question of how they are derived.

To get there is hard but once the work is done, it’s in the background and becomes incidental. Moreover, I would have more faith in something with a theoretical basis rather than some TA indicator which is the result of a bad trip by some guy who looks “connected”. Yes, I do use TA (I’ll also look for the Vegas method).

I’m always looking to reduce uncertainty re market direction and size of move. And the greater the time period, the greater the uncertainty. In other words, I can never predict market direction, nor the size of any move. Therefore, I consider scalping as appropriate to my way of thinking, and my limitations.

“different slant on things and get an edge”. What I’m looking at from a theoretical point of view is standard (in options theory), old hat. It is far from original. How often are key levels in fx cited as being heavily protected/ attacked because of some knock-out, knock-in or barrier option, or some other exotic? It may be incidental, but occasionally it’s significant. In the context of index futures (my area of interest) a sudden jump in implied will usually precede a sell-off in futures (hedging). Basically, your looking at an outcome from two, rather than one perspective. To “trade on price action” is the same as trading cheap or expensive volatility.

My edge is reducing My risk but maintaining the same profit potential.

Watch out for sun burn, comrade.

Grant.

 

[kiya]

i've just started to read up and find out about trading.I'm not to sure what area i should concentrate on to learn.I'm prob not ready to start trading yet but i dipped my foot in with 3 companies and due to the Chinese the market took a big dip and my shares took a slight dip as well.I was gony hold onto them in the hope i can at least break even or does any one think i should sell and cut my loses for the time being.
KIYA

 

[pssonice]

yes, the larger your time frame the shorter your moving average

u have to use the weighted moving average to get rid of the influence of obsolete data