Thanks for your comment - I’ve just checked this in my platform and you’re correct, the times shown are indeed New York Time. I’d failed to switch my charts from exchange time to local time (I’m in the UK) when setting up the analysis, which is rather a clumsy mistake.
It’s interesting to see the level of similarity in the 2011 chart you’ve linked to, as there’s only a year or so of overlap in the data used. It’s also made me curious enough to think that it’s worthwhile adding in volume for the remaining studies.
London session, from start to end is pretty good to inter day trade or scalp. As it is the biggest session during a trading day so you can make good pips as the movements are healthy in this session.
Thanks for sharing these! I prefer the session overlaps myself for quick trades and hunting for pullback opportunities. Do share the other pairs you’re looking at as well!
This week I’ve looked at USD/JPY. Depending on who you ask, it’s the second most traded pair after EUR/USD.
I’ve added in some volume data alongside the volatility for anyone who is interested; there’s evidently a strong positive correlation between the two. I wouldn’t read too much into the dip in volume around the US close, as the volume data has come from the full sized JY futures contract which trades on the CME, and everything gets a bit patchy around this time when Globex shuts down. It is interesting to note that the futures price does continue to track the spot price during this period though, which suggests to me that very little volume from futures/spot arbitrageurs is actually require to keep movements in the two in line when there is less speculative volume.
Anyway, the volume data is vaguely indicative - don’t expect that it’s especially representative of volumes in the interbank market.
Another point to note is that the secondary peak around the East Asian open shows just as much activity for this pair as the London open (unlike with EUR/USD), which might be taken to imply that pairs are more heavily traded during their “home” session. I’ll run with the British Pound next week, and we’ll see whether this confirms that idea.
I don’t personally believe that it would be possible for any of us to extract any meaningful returns from these volatility relationships themselves (although you can imagine some options trading high frequency firm with ultra low costs pulling of some kind of volatility stat arb - long volatility for one pair and short volatility in another in an attempt to remain delta neutral) - the cost would kill even if the theory played out. My hope is simply that they provide some guidance on the best times to trade for those of you that enter and exit positions intraday. Of course, this is completely strategy dependent - if you’re trading mean reversions then these might be the very worst times of day to trade!
Though the patterns we see here aren’t completely different to the results of our Yen and Euro analysis (each of the three international forex trading center opens is accompanied with pronounced volume and volatility), there are a few noteworthy differences.
The greatest volatility is seen, as we thought it might be, following the London open. The volumes accompanying this volatility peak, however, are nowhere near as significant as those we see in the US markets later in the day. This is interesting, as low volumes can be associated either with a lack of speculative interest in a forex pair and a quiet, ranging market, or with a lack of liquidity and volatile behavior. If the market is thin then an influx of one sided interest can cause price spikes and stop runs.
Moving later on into the trading day we can see that volatility appears less pronounced in the Asian session than in our other studies. Traders in the Far East are less likely to move the British Pound, it seems. This could be useful information to keep in mind if you are considering holding a position in GBP/USD overnight following the close of the US session: on average you’re not likely to have to endure significant volatility until the London open the following day (obviously, exceptions will apply!).
As always, if anyone has any requests, questions, suggestions, or you’ve spotted errors in the analysis, then let me know!
I played around with your first chart (EUR/USD with price histogram only) to see what it would look like if the 3 major sessions were shown. Here’s the result —
I think that showing the trading sessions adds to understanding the time period that’s being shown — but, do with it what you will.
I considered labeling the all-important London/New York overlap period, but thought it made the chart look cluttered.
Explanatory note: Two different 9-hour sessions are shown for Tokyo, depending on time of year. At first glance, it may appear that Tokyo is going onto and off of Daylight Saving Time (which is not the case, as Japan remains on standard time year-round). Using New York time as the x-axis (as you have done), times in the U.S. and U.K. [I]appear not to change [/I]throughout the year, while times in countries that do not observe DST [I]do appear to change.[/I]
[QUOTE=“BestBrokerDeals;728618”]I’ve just started a weekly series over on the website where I’m going to investigate the best time of day to trade particular markets - I thought I’d post the research focusing on forex pairs over here as well, as any intraday traders might find them useful. What I’m going to be doing is really just a very simple a study of when the markets are most volatile - the times when the market is most likely to provide large intraday movements. Each day is broken down into hours - these are what you’ll see along the bottom axis (the time shown is the close of each hour period, so the bar labelled ‘5’ represents volatility in the hour between 0400 and 0500). All times are EDT (New York Time). I’ll be going over all the major pairs over the coming weeks, but there might be gaps in posting as I won’t post in weeks when I’m focusing on a commodity or index rather than a currency. This kind of statistical analysis can be incorporated into a discretionary strategy or form part of quantitative and automated approach. If anyone has any particular requests for pairs to cover then please post them in this thread. Nick[/QUOTE]
Anytime where you can get it and get out with a profit. However that applies to your situation.
Best time is when your fully awake and your head is loaded 100%.
aside that the guys who postet before me showed good tables and graphs onto when to trade and when not.
Theyre right. I never used graphs, i used common sense from real life to understand after a week when theres action when there istn.
it rather easy as we are all human beeing we all follow the same patterns.
High volume
Openings of major markets. Frankfurt stock exchange opening 09:00 (london time 08:00, new york time 01:00)
Opening of London - (Frankfurt time 10:00, new york time 02:00)
Opening of New Yrok mrket - (Frankfurt time 15:00, london time 14:00)
Low volume
Midday lunch - Frankfurt (London time 11:00 - new york time 05:00)
Midday lunch - london (frankfurt time 12:00 - new york time 06:00)
Midday lunch - New york (frankfurt time 18:00 - london time 17:00)
Then the pre closening times of the stock exchanges, half a hour before closing you can expect a increase of volume as the deciders have observed the daily action and onto it make their decition about the newx few days of their idea where the market is going to go.
yes youre right, i apologize. I should have explained in more detail what I meant. I was thinking of my own opening hours the computer trade of xetra as most banks act upon private traders on the xetra times and the banks itelf trade on these times.
90% of the trades in germany happen on the xetra computer exchange. their both the same stock exchange but have different opening hours. <- german logic, weird i know.
The volumes in frankfurt still come it at 09:00 (8 GMT) even if the old fashioned parkett based frankfurter börse opens at 08:00 (07:00 GMT).
Quote Xetra explanation:
“Im Jahr 2015 wurden über 90 Prozent des gesamten Aktienhandels an deutschen Börsen über den Handelsplatz Xetra abgewickelt. In den DAX-Werten erreicht Xetra europaweit einen Marktanteil von über 60 Prozent.[4] Die Handelszeiten am Handelsplatz Xetra sind börsentäglich von 9.00 bis 17.30 Uhr.[5] Der Präsenzbörse der Frankfurter Wertpapierbörse wurde in Frankfurt am Main am 20. Mai 2011 auf die Xetra Handelstechnologie umgestellt. Die Preise auf Xetra sind Basis zur Berechnung des bekanntesten deutschen Aktienindex DAX.”
Short translation:
90% of all trades in germany happen over Xetra.
The DAX worth, point-count, is calculated upon Xetra numbers only.
Nobody would argue the fact that [B]stock-trading sessions[/B] in each country begin and end when stock exchanges open and close in their respective countries. But, spot currencies are not traded on stock exchanges. Spot currencies are traded off-exchange, 24 hours per day, 5 days per week, around the world.
[B]Forex (spot currency) trading sessions are not defined by the operating hours of stock exchanges.
Forex trading sessions are defined by the daytime operating hours of major banks[/B] —
— in three principal locations: Tokyo, London, and New York.
Those daytime operating hours are generally assumed to be something like 8am-5pm, local time, in each location.
•
You and I have run down this rabbit-trail once before. Here is the reply that I gave you last time:
Clint is right, as usual from other posts of him ive seen.
But theres one thing i cant withstand to add.
Yes, again common sense, most trading beeing made while the people are awake. Forex isnt bound by times like stocks, but bound by habbits. Has anyone of you ever treid to trade in your lokal time of 03:00?
Theres really no need to think to much of this topic at all, its a time waister, common sense tells everyone, starting from 09:00 to 17:00 when the instituionals are trading is the time the biggest chunk of contracts is beeing made. instituionals finish their job, at 17:30 their home, they dont care what EUR/USD isdoing untill the next day 09:00 when they arrive at their working place. in everyones local time. But not only isntitutionals, pretty much everyone knws from birth that working time is from morning till evening, after that its hobby/family/restaurant/tv-time.
except maybe if theres a vampire country of which we still dont know of its exstance, in that case sunset would make a good volume phaze, midnight is “we look for nice girls to sucke their blood out”-time and morning is “ah lets watch morning tv and go back to sleep”-time.
theres really no need to go complicated with graphs, studies, etc. over such a common sense topic. no insult to anyone, if someone feels insulted it wasnt my intention. but i think simplicity helps a lot in many things in life, trading aswell.
Thanks for that - adding the time zones makes it much clearer and is a great suggestion - I’m going to go back into the original posts and edit them to include this (although I’ll probably show only the full possible span of the Asian session to keep things uncluttered).
Yes, agreed - it’s all strategy dependent - if you’re using a market-making strategy then typically volatile periods might be the very [I]worst [/I]times to trade!
Once I’ve been through all the major pairs I’ll try and put together a post looking at how these volatility patterns affect the results of a couple of example strategies in different styles.
I’ve come unstuck many times as a trader by relying on “common sense”. I’ve also been surprised over and over by the degree to which patterns in price data perturb expectations. Over time I’ve learned that the most fool-proof (and even “common-sense”) approach is to take nothing for granted and interrogate the data.
One main reason that algorithmic trading has come to dominate the market (not only in the high frequency sector) is because it doesn’t bring the biases and make the often incorrect “common-sense” assumptions that human traders do.
A final point - I wouldn’t say there’s anything “complicated” or overly sophisticated about the analysis I’m posting - we’re not trying to unpick deviation in theta decay of an in-the-money option call hedged with a dynamically allocated basket of triple-leveraged ETFs . . . or anything silly like that - it’s just a bar chart with a very basic volatility measure.
There is. It’s called finance . . . On every exchange, lurking behind every dealing desk and gaming every ECN are vampires. They want your blood and your money and, to misquote Gary Oldman, they’ll cross oceans of time-zones to find you!
above i was talking bout common sense in human behaviour more then of common sense in markets, i know markets dont act upon common sense very often and your right about what you wrote, we only talked about 2 different things
With oil proving to be one of the principal drivers for markets so far this year, it seemed like a good time to tackle the major commodity currencies, starting with the Australian Dollar.
The AUD/USD pair appears to be notable for lower deviations in its volatility throughout the trading day. Although it is possible to identify the three characteristic peaks in volatility accompanied by higher volumes, these are both far less pronounced when compared to the troughs, and far less different from one another. The Australian Dollar perhaps has less of a ‘home’, and is consistently actively traded around the globe.
While volume is notably higher around the start of the US trading day, it should be borne in mind that some of this is due to the natural influx of liquidity at the start of the regular futures trading session.
Although the Australian Dollar trades with the fifth highest volumes of any currency (due in part to international trading of natural materials, and due in further part to the carry trade), this apparent absence of any predictably enhanced period of volatility suggests that it is possibly not the best choice for daytraders who require significant directional price changes.
