When you have the time, could you explain how you calculated the > or < 3 R/R?Originally Posted by grems8544
If I take IWM as an example, are you using the Resistance to Support ratio which is 2.089 currently?
Thank you
When you have the time, could you explain how you calculated the > or < 3 R/R?
If I take IWM as an example, are you using the Resistance to Support ratio which is 2.089 currently?
Thank you
GGT Forum Manager
The first step is to calculate all of the pivot levels from the daily out to the yearly. Additionally, I calculate the simple moving averages for the 50d and 200d. Per Billy's and Maxime's methods the following weights are assigned:
Yearly: 8
200d: 7
Semester: 6
Quarterly: 5
50d: 4
Monthly: 3
Weekly: 2
Daily: 1
To calculate R/R, Maxime's method searches from the present price upward to the first pivot that has a weight of 3 or higher. The distance between the present price and this pivot level is determined and is viewable in my spreadsheet in the "Dist2Res" column.
In a similar manner, the distance between the present price and the lower support level is returned. This is found in the "Dist2Sup" column.
The R/R ratio is simply the result of Dist2Res/Dist2Sup.
This is Maxime's method and as a baseline, it works well.
A few comments:
1) Dis2Sup can be a very low number as the price drops to the support level. In fact, R/R can sky rocket to >>3 as the difference between the present price and the first support level collapses to 0. This is a non-linear behavior that bothers me because the R/R value becomes disproportionately large as the difference value in the denominator tends towards 0.
2) Dis2Res can become 0 as our prices approach the resistance level from below. Same problem with non-linearity as above but now we tend to 0 in the R/R, which is not completely accurate.
I am working on an algorithm right now that uses the cluster weights instead of the the absolute pivot levels. This seems to give a smoother transition of R/R as we move around within clusters, and it will better for computerized trading. The impact of this weighting is revealed in the column "Bal", which represents the cluster weighting "balance" on the resistance size compared to that on the support side. A natural hypothesis is that stocks with very little overhead cluster weighting should appreciate more than stocks with heavy overhead cluster weighting. Here are the stocks with the most favorable cluster weighting going into the open on Friday morning:
Correspondingly, here are the stocks with the least favorable cluster weighting:
Work in progress...
Regards,
Paul
Paul, this is terrific data. I have been relying on Billy's pivots and R/R which are great but limited to IWM and GDX. Thanks to Maxime, I have been calculating the R/R on other stocks using Billy's guidance (delta between 1st strongest resistance and 1st strongest support). Do you plan on combining and publishing your leading stocks along with the PP balance on a regular basis, or should I assume that your leading stocks have a favorable PP balance?
I would be lost without the support of guys like you, Maxime, Billy and Pascal.
My best,
Buzz
GGT Forum Manager
No, do not assume this. I'm not comfortable selecting stocks that have a favorable RR balance and offering it as guidance simply because I've not tested it. It *appears* to have an edge, but I do not have any statistics to guide. From a gut-feeling point of view, and with a bit of knowledge about supply/demand, stocks with little overhead resistance should have very little supply (e.g., those folks who have been holding at those higher levels at a loss will be willing to unload the moment they are at the water line and even. Note this has a theme very close to Pascal's AB concepts.), and with little supply available, we should see price appreciation.
The problem here is that we have a correlation issue. Simply put, some stocks react far more in line with the broader markets than do others. This makes the decomposition more difficult, e.g., it is harder to discern movement due to pivot resistance due to broad market movement.
My internal work is focused on principle component analysis, or PCA. Simply put, given a series of prices from several indexes as well as the equity, we should be able to start by removing, or detrending, the price series in a method that removes the influence of the index. Hence, if we start with the highest beta component, which generally is the equity price series, we can run the series through a correlation calculation to determine which index is most highly coordinated with the equity. Once we know this, we can apply PCA transforms to remove the influence of the index, and what remains (in principle) is now the stock behavior alone, without influences of markets. Note that it's impossible to do this in reality, because if you change the starting date of the series, you get a completely different set of weights which may be close to the previous set (or not). This is the part I'm struggling with right now. The implications are that two data sets, 500 days long, offset by 1 day, and run through this PCA mumbo-jumbo magic, do not produce curves that virtually overlap each other. This is the specific problem that I'm trying to understand. The magic works on canned data that I created, so I know I'm close, but not close enough.
Provided that I can overcome this pain-in-the-neck technicality, the next step is to recalculate all of the pivots using a detrending PCA algorithm. Basically, think of the pivots table that Maxime has constructed, but now do so without the influence of the markets. Once I have this, I can then test the assertion that stocks with little overhead resistance have a greater chance to move higher.
Regards,
Paul
Sounds like a worthwhile project. While watching for the outcome of your future reseach, I will continue on the path I'm following today, that is, take the path of least resistance. Intuitively, it makes sense and moreover with the use of Maximes PPs, easy to determine. Whether the stock advanced due to market conditions or least resistance is secondary to me at the moment; I simply want them both in line in the same direction. If you are able to validate that there is a strong correlation between a stocks' gain and your low balance number, you will be making a major contribution. You are a great asset to anyone serious about trading. Keep up the good work, and lol.
Regards,
Buzz
GGT Forum Manager
Thanks Buzz. Therein lies the issue though -- you want them to both align (the movement of the stocks and the market). This is only beneficial if it occurs AFTER you have bought the stock -- watching a 100 stocks move in the direction of the market but not being in any one of them does you no good.
At the core of this is the question whether pivots offer an edge. Specifically for us: if I select stocks for entry tomorrow which have virtually no overhead resistance, do these tend to go up in a higher percentage compared to a random selection of stocks on the same day? Furthermore, if I select these stocks and tomorrow is a DOWN day and the stock drops in price, is it because pivots played a role in the selection (meaning I screened on pivots and they are biased to the down side)?
The best outcome is one where low overhead resistance outperforms a random sampling of stocks on any given day.
Thanks for all of your feedback. Make it a great weekend.
Paul
Paul,
Thanks for this.
To use a sports metaphor: I'm now fearful we'll lose you to free agency!
Best,
Paul, in calculating the r1, r2, s1, s2 clusters, what ATR multiples do you use? bz
GGT Forum Manager
+1, +2, -1, -2.
This is fascinating work in progress. I especially like the cluster weighting balance idea which I think has a lot of potential. In my research, I use the median of the Billy support and resistance clusters for each stock in my database in order to "support" my evaluation of S and R with P&F charts..
In any event, I believe I asked the right question this time based on other comments.
You are invaluable... A real research "machine" [lol] and I appreciate the long hours you are "giving" us day after day
Posting Permissions
Reply With Quote