I hear some people say: “because traders etc. are off on holiday (school summer holidays too etc.) from July – August and then September everyone is back on it”…
I’m not sure what markets they’re referring to, but let’s have a crude look at EURUSD price data to substantiate whether September is more “volatile” in comparison to the rest of the months in the year:
What I’ve done is downloaded (relatively high quality data too) EURUSD H1 time frame data from 2001–2016 Sep (which is 93,000 rows of data in excel).
I’ve then done the Log returns of the close prices (excel =LN(p/p-1) – where “p” is the period) and then got the Std Deviation of each Log return on the H1 for the corresponding month (I also transformed the LN into a percentage with (e^ln-1)).
Ok, what… i don’t have a clue what that means!?
If you’re not into maths (or if you are, don’t get too picky with how I’ve done this crude test!), the graph above proves that every month of July and August since 2001 to current seem to display a lower standard deviation (measurement of volatility) than that of all the Septembers!
Unfortunately I cannot explain nor give you a fair response as to why markets are “volatile in September” just as I cannot explain to you why January, June, October, November and December are more volatile than September (at least on EURUSD). However, the theory that traders are ‘off’ during July and August (particularly August) would definitely be true just by looking at the market information.
What’s interesting is if you considered how dramatically different the change is. Take the drop from June to July and increase from August to September. They’re quite significant and it’s definitely a telling sign about how the market behaves during these periods in the past.
August is certainly a very low volatility month to be trading and you’d definitely want to consider transaction cost relative to opportunity.