Portfolio Boss Documentation

Average Percent Range (APR) Filter



This filter calculates the volatility of an instrument during a certain period, and if such volatility falls within the set criteria, a Buy (or Sell) signal is given.

Unlike other volatility filters, the volatility here is calculated similar to the Average True Range (ATR). Whereas ATR gives the volatility reading in dollar value (how many dollars a given instrument moves per day, on average), this filter converts that dollar value into percent.



The idea is this: volatility stated in dollar value is very subjective. An average movement of $15 per day might not be much for a stock priced at $1000, but the same $15 movement for a $300 stock is considered very volatile. Therefore, converting such volatility into percent (relative to the instrument's price) allows for a more objective judgment. This is especially true for Portfolio Boss, where hundreds (if not thousands) of instruments with different price range are calculated at once. Even the same instrument could have different prices across different periods. Thus for our purposes, we believe Average Percent Range (APR) gives a fairer volatility reading than the Average True Range.



1.  The first parameter defines the lookback period for the volatility.



For example with a period of 50 days, the average volatility is calculated from 50 days ago until the latest date (today). If today's APR thus falls within the threshold APR, the buy (or sell) signal is given.

Generally, a longer period is preferable (upward of 50 days), so the volatility reading is more accurate. A shorter period may not smooth the outlier volatility enough (temporary spike in price), hence giving false signals. But there are cases where both a longer period and a shorter period are used at once (with two filters), to exploit a temporary divergence in volatility.



2.  The second parameter defines where the instrument's APR must fall (in relation to the threshold APR) for the signal to be given.



You can choose here, whether the instrument's APR must be “Greater Than”, “Less Than”, “Between”, or “Not Between” the threshold APR. Generally, a Buy Filter is best set to “Greater Than”, while “Less Than” is good for a Sell Filter. The reason is that higher volatility gives higher return, so you buy (but beware of higher risk too). And when volatility is low, it could be that price is trapped in a trading range.



But sometimes the reverse is true (in a primary bull market, for example): price tends to drop much more violently & quickly when volatility is high, while the uptrend is a slow grind with lower volatility. So, set this parameter according to your strategy's characteristics.



3.  The third parameter defines the threshold APR, where instruments' APR are compared against.



Instruments' APR are usually less than 15%, so set this parameter to such a small value (3, 6, 7, 14, etc). A high APR threshold (like 50% or more) gives too much leeway for the instruments' APR, hence inaccurate signals. Besides, such a high APR is rarely encountered, so the strategy may stay in cash most of the time.



Sometimes, a volatility indicator by itself is not enough: it doesn't take into account whether price is going up or down. So it's a good idea to supplement this with another filter. For example, aside from the spike in volatility, another filter is used to identify a young uptrend (or the end of a downtrend). Hence the high volatility represents investors' eagerness to take the price even higher, instead of panicked selling:



In the example above, the Aroon Oscillator (ARO) Filter is set to identify the end of a downtrend (i.e “Above ARO -60”). Then coupled with the increase in volatility (i.e “APR Greater Than 6%”) we have a confirmation of a strong upward momentum.

You can even use this APR filter twice: one Buy Filter uses a shorter period, while another Buy Filter uses a longer period. As stated somewhere above, this technique is used for exploiting temporary divergences in price volatility:



In summary, you can pair this filter with other filters as you see fit, building upon the general ideas set out in this Chapter 7.





  • As stated above, the APR calculation is slightly different from ATR (aside from the fact that ATR doesn't yield percentage values). Unlike ATR, APR does not take into account the previous day's close (just the difference between a day's lowest and highest prices). This is the correct way for stock and ETF prices. Because ATR, on the other hand, was originally developed for commodities' futures prices.
  • At the Instrument Tab, there's an APR indicator below the price chart. Change the APR Filter's period parameter to change this indicator's appearance.






Back to Top