Choppiest pairs- Most Volatile within Range

joseclar

Newbie
Messages
7
Likes
0
I have been doing a lot of research on volatility. What would your approach be to determing the volatility within the daily range?. For instance say the daily range on the eur/gbp is mid 80s and the daily range on gbp/jpy is high 200's, what techniquies would you use to determine, how many times within that high and low range the pair moved 10 pips, 20 pip, 30 pips in on direction and so on. I suppose some form of regression could work. I could all use a brute force analysis but that seems very time consuming. Please let me know what you think. If you can think of any indicators or statistical analysis techniques that could give me the data I am looking for. Thank so much for the reply.

Basically I am looking for your opinion on what pairs are most choppy, meaning most likely to retrace frequently during their ranges or trends, and what techniques you would use to evluate this special form of volatility. It seems that some of the cross pairs are pretty choppy. I have been looking at nzd, aud, eur, and gbp crosses, as well as majors.

I would love any input to my overall question. But most importantly I would like suggestions on pairs that are choppy. Cross pairs are fine. I am looking as mentioned for pairs that wont go on large single directional runs, with out retracing. Thanks.
 
ATR doesnt come close to what I am looking for. Thank you for the reply.
Daily high lows, moving averages, atr, can help in reflecting overall volatility or common daily ranges. I am talking about volatility within a range.

So if on any given day a pair moves 100 pips in a given direction, how much actual movement was there. How sharp or choppy was the trend? How many times did it move a significant amount within that daily range?
 
Actually, you could do something like taking the intraday ATR and comparing it to the day's range. If you multiply out the intraday ATR by the number of periods you'd get total ground covered during the day's span.
 
I suppose you could load a tick chart than then code an indicator to count the number or ticks per time interval... Then you code another indicator to tell you the total movement during the same time interval. Then work it out from that?

Edit: Or do what RT said ;-)
 
Thanks. My best attempt so far has been using various regressions on the smoothing data like MAs. I like the multiple ATR idea. Can you clarify how you would do it. I dont have much experience with ATR. When you say intraday, how do you set it accordingly? Thanks
 
Assuming an 8-hour trading day, you could run an 8-period ATR. That way, at the end of the day, you have the average hourly range for the day. Multiply that by 8 and you get the amount of ground covered intraday (of course you could use shorter or longer timeframes if they better suit your needs). Then you can compare that to the day's high-low range.
 
And then comparing that to the spread might give you a more accurate idea of tradeable moves.
 
Top