Portfolio Boss Documentation

All-Time High/Low Filter

 

 

This filter allows the strategy to enter or exit Positions if an all-time high (or all-time low) has been reached within the specified period. All-time high (ATH) is defined as the highest adjusted closing price ever reached by the instrument; and all-time low (ATL) is the lowest adjusted closing price the instrument ever hit throughout its lifetime.

 

 

  • Generally you can use this as a Buy Filter to look for all-time highs, since these indicate strong bullish sentiment. New highs are usually brought about by some groundbreaking favorable news or circumstances for the company itself, or the market in general, and these will attract droves of greedy investors who will drive up the price even more.
  • You can also use this as a Short Filter that finds all-time lows. Companies (or funds) that post a new low scare off investors, because usually there's something fundamentally wrong with them. And it would be hard for their stock price to recover.

In other words, you use the filter to enter Positions. It's not recommended to use it as a Sell (or Cover) Filter, because by the time the all-time low is reached it would be too late to get out as your account size has gone downhill for the most part. Theoretically you can (use it to exit Positions), but just make sure your entry is also close enough to the all-time low (or high, if shorting).

 

 

Now, some mavericks out there see all-time highs as an opportunity to short (investors are “scared of the heights” they say), but this is more of a folly than wisdom. You short only when you see weakness, not strength. But just try it and experiment for yourself, maybe this out-of-the box “wisdom” could be the secret sauce for your strategy.

All that being said, it would be prudent for you to set other criteria for entering the Positions. All-time high (or low) by itself is still a very loose criteria, it may be good to wait a little before jumping onto the bandwagon (who knows, maybe the bulls need to take a rest in a shallow creek up ahead). Or to see whether the momentum (and volume!) are still strong indeed to warrant further hikes:

 

 

Don't be part of the crowds gullible to overhyped news. How many times did investors chase really hot stocks, only to realize later it's the peak!

Now, let's discuss its parameters.

 


 

1.  The first parameter (leftmost) lets you define whether the filter should find all-time high, or, all-time low:

 

 

The general wisdom says to find all-time highs for a Buy Filter, and for the Sell Filter you want to find all-time lows. But you can't go wrong with the opposite too (as our tests indicate, helped by the Divine Engine); that is, set this parameter to “Low” for a Buy Filter, and “High” for a Sell Filter. It all depends on your strategy's rules, filters, and Portfolios.

 

 


 

2.  The second parameter defines the lookback period to find the all-time high or all-time low:

 

 

This obviously includes today's candle (once the market closed). Let's say you set this parameter to 5 days:

 

 

then the filter must identify such ATH or ATL within the span of 5 days, including today. Our tests indicate a short period from 20 to 55 days to be generally good, as well as a longer period of 110 to 160 days. Side note, the minimum value you can set here is 2 days, which means you jump into the ATH or ATL (nearly) as soon as it happens. If you want to “wait a bit”, set a longer period here.

With longer periods, sometimes the same ATH (or ATL) can trigger multiple (subsequent) signals.

 


 

Notes:

 

  • The new all-time high does not necessarily be higher than the previous ATH; it can be exactly the same price (but not lower). Ditto all-time low, i.e. it must be equal to, or lower than, the last ATL.
  • The all-time high (or low) is compared to the previous prices all the way from when the instrument was first traded (listed) on the exchanges, not from the earliest backtest date of the strategy.
  • Since this filter is all about the instrument's price, you may be wondering why'd we use adjusted prices? For backtesting purposes (which Portfolio Boss is all about), adjusted prices is the most realistic price. Imagine if due to a stock split a company traded from $100 down to $5: while such a low price may be the lowest price it ever traded (and even breached all support and moving averages), it's not an indication there's a huge selling pressure or that the company is about to go bankrupt. It would be wrong to base our technical analysis on that faulty “real price” assumption (as it's simply a stock-split, or dividend payment, or what have you). Now don't worry, because when you finally trade with real money (that is, forward performance as opposed to backtesting), the strategy will definitely use the real price, as adjustments happen after the fact, not before the stock splits happen.
  • Some market technicians advocate the identification of multiple consecutive ATH (or ATL) before placing a trade, as a way of confirmation. Due to the way this filter is calculated, you can't find such consecutive ATH/ATL even if you stack this multiple times as Buy Filters (with different periods). The shorter period will always trump the longer period filter in finding the ATH/ATL.

 

 

 

Back to Top