the problem that I have with their method is they aren't applying a buying/selling scale factor to the total volume. Never will all volume be strictly buying OR strictly selling.
I have playing some more with it, still nothing formal, but I was trying to see if I could somehow give weight to the formula in 'who' is in control. Remember we are talking about a 'L/R - liquidity ratio' that determines how much volume is needed to move a stock a point; based on the originator's definitiion.
I found that by playing with it, the spikes just make it impossible for this to be useful in a 5-day format. I mention 5-day because to calculate their formula, it looks as though I have to calculate everyday to find the percentage and gave up after 5 days for testing the formula. They use 1 month of data.
What I was thinking, others jump in here, is can one assume certain things based on today's O/H/L/C and yesterday's. Been trying something like this:
I like binary systems & indicators and then use the indicator's value for reference;
Open >= Yesterday's Open --- Buyers In Control 20 Points
High >= "" "" "" High --- Buyers In Control 20 Points
Low >= "" "" "" Low --- Buyers In Control 20 Points
Close >= "" "" "" Close -- Buyers In Control 20 Points
Typical Price >= "" "" Typical Price -- Buyers In Control 20 Points
I was using 10 different things and 10 points each to make 100%, but let's use the above. I am assuming that if Today's Open is greater than yesterday's, then there are buyer's etc.. Now if the total is 100, then 100% of the volume is used to indicate an up day. If the total was 40, then 40% of the volume would be used for the upside and 60% of the volume would be used for the down side and the total of the indicator would be the final product.
I may not be making sense on the indicator as I am shooting from memory here and playing with the L/R formula I thought looked interesting. I am still working on it to see if I can make something 'useful' out of it. But concerning trying to really determine the volume on an up or down day, would the approach have some kind of merit? Any opinions anyone?
I think what I am trying to do is breakup the volume for the day so that of 100,000 shares traded for the day and using the above percentage points for how the day is compared to yesterday, would show how much volume is happening towards the upside and how much volume was on the downside. So it could turn out using something like this that down volume was broken up to be 40% while the up volume was 60% for the day.
Then maybe these 2 numbers can be used to create other indicators. Another reason I am playing with breaking up volume is to plot them both and see if any crosses show anything useful or average of the 2 bring anything useful, etc... I find the challenge interesting to see if I can do it.
You mentioned that the L/R I saw didn't account for volume being up/down or something like that, which started me on this, and it is fun. But I think I also am liking the work involved because in Wow Pro, I don't know of any indicator that shows the breakup of up and down days. At least not that will show me in code so I can understand it. I would prefer to have a code of High-Low = Something and know what it does versus Fml("Special Formula") >= Ref(Fml("Special Formula"), -1) since I don't know what Special Formula is, although it might do the same thing. Code is nicer to understand than the canned indicators. Most of the indicators keep saying, In So & So's book, here is the formula. Well, I don't have the book so don't know how it was calculated and maybe I would like to change it.
I think I am rambling now.... <g>
Tim |
