Williams %R
suggested by Ken G.

When buying and selling stock the principle rule is to Buy Low, Sell High.

>Brilliant! Is that idea yours?
Pay attention.
The questions become:

  1. Low compared to what?
  2. High compared to what?
There are a jillion suggested answers to these questions, like Bollinger bands, moving averages, MACD, stochastics, Fibonacci, etc. (See this.)

One that I never heard of is Williams %R (named after Larry Williams). I was, however, surprised to find that it's just the stochastic oscillator ... but upside down?

>Huh?
Williams %R says:

  1. Low compared to the lowest price over the past N days.
  2. High compared to the highest price over the past N days.

Williams %R
 		Williams measures the current price on a scale running down from the highest to the lowest price.

      %R = 100(High - Price) / (High - Low)

  1. When %R is close to 100%, the price is close to the Low over the past N days.
  2. When %R is close to 0%, the price is close to the High over the past N days.
The answers to the above questions?
  1. Buy when %R is close to 100%
  2. Sell when %R is close to 0%
>And that stochastic thing?
That puts the current price on a scale running up from the lowest to the highest price (over the past N days).

      %K = 100(Price - Low) / (High - Low)


Stochastic oscillator
Notice that Buy and Sell signals are the same. Only the name has been changed.

>Buy and Sell signals?
Yes. When %R is less than, say 20%, then Buy and ...
>When it's above, say 80%, then Sell. Right?
You got it.
Here's an example (using daily General Motors prices over the past few months)

There are:
      20% and 80% levels.
      A plot of %R and the stock Price.
      Whenever %R drops below 20% there's a Sell signal.
      When it rises above 80% ...
>Sell!
NO. It's Buy.


Willimas %R for Microsoft

To download a .ZIPd spreadsheet which does this sort of thing, RIGHT-click here and Save target file
... but don't expect miracles