How to determine whether today's session will be high or low volatility

Discussion in 'Trading' started by rolando87, Jan 24, 2013.

  1. I figured that my trading system would benefit if I can somehow figure out (before the trading session begins) if the upcoming trading session will be a higher than usual volatility day or lower. I'm not asking for a crystal ball method just some methodology to "guess" what today might bring in terms of movement in the markets. Direction is not important, just how much of a range I can expect to see over the day. I mostly trade bonds, and eurodollars, but if the S&P will move hard then I expect the bonds to move too. Everything pretty correlated nowadays anyways.

    The things I'm currently looking at to determine this is taking a look at all scheduled economic numbers that are due that day. Taking a look at major indices to see if we are likely to break new highs or lows, or whether today's action is contained within yesterday's daily bar. And any major news.

    Anyone recommend anything else? Maybe some websites that aggregate this type of data?
  2. Because your system does well with volatility.. and gets killed with noisey days...?

    Volatility clusters.. vol begets more vol.... other then the obvious event trading ... I'd stay away from Fridays before holidays and triple witch...
  3. my strat is exactly the same, I would do just fine, long or short, if I knew what the ADR was going to be. But wouldn't we all?

    rather than guessing and predicting

    I'm working on a way to react

    It's always a little late, what they call a "lagging indicator"

    but if you have a long term strat, it evolves over time

    whether that time is a few minutes, or a few years
  4. My system dose fine in both high and low volatility days but I would like to adjust my size accordingly to take on smaller size on days where we might see a bigger range, and jack up my size on low volatility days. That's basically where I'm coming from.

    Isn't there any news sites that specialize in predicting this or at least come close. Thanks
  5. If you figure out how to do this, let me know!

    Would make trading easier. If you knew ahead of time how trendy it was going to be, you could trade with something as simple as MACD, and just stay out on the choppy days.
  6. Well in that case anyone tell me how to buy the low of the day and sell the high? It wil lhelp me a LOT :)

    PS: It can't be done.
  7. 1) You could use the "sigma calculation" from the implied volatility of the options as a way of detemining an expected daily range for a market. Based on the opening high & low in the market you can anticipate what the daily high & low may be later in the session. :cool:
    2) It's not a guarantee of exactly forecasting a market but the accuracy may pleasantly surprise you. You will still have to contend with intra-day price movement that may prematurely shake you out of trades before hitting an expected price target. :)
  8. carlos99


    I will take a look at options. Thanks for that suggestion. In case there is any confusion, i'm not trying to determine whether the day will be choppy or trend in one direction or another but to determine if the high - low will be higher or lower than average. A choppy day could have a larger range than a trending up day.

    I was also thinking floor trading pivot points can help in gauging approx range of the day as I've noticed price action in the spx will gravitate towards the pivot points. At least between s1 and r1. Bollinger bands too or something like them that uses the preceeding days action when calculating the standard deviation lines.
  9. From a time series and econometrics perspective, you might want to look into various volatility estimators (GARCH, etc). Estimating volatility or squared/absolute returns are much more reliable than directional forecasting.

    As someone posted earlier, 'volatility begets volatility,' and volatility exhibits heteroskedasticity clustering.
  10. i think it is helpful to distinguish between volatility and range.

    Range = how far price travels from high to low.
    Volatility = how much movement there is within that range; how much backing and filling etc.

    Range is very difficult to predict. Outsized range usually seems to be due to unexpected news.

    Volatility is easier to predict. Expect volatility to be about the same as it was yesterday. Low volatility markets tend to remain low volatility until shaken by some kind of news. High volatility markets tend to stay highly volatile for a while and then gradually become less volatile.
    #10     Jan 27, 2013