RSI in Laguerre Time with Fractal Energy - Momentum/Trend Indicator

markos

markos

Well-known member
VIP
@markos Hi Markos.
I have a Laguerre question for you simply because I need validation.
I have pulled up a BYND daily chart w Laguerre. Looking at the Laguerre, RSI at the bottom for a long downtrend, FE on 10/21 is showing drying up of this trend. I assume the Laguerre takes a few days and about now (although with earnings, not"now") but let's say no earnings, here with these last two candles seems a bottom or if there were no earnings at least the beginning of consolidation. What say you? Is this assumption correct?


Hi @MBF sure, i'll explain what I see, I tend to ramble when it's late, please try to fill in the missing pieces by asking a follow up question. Your assumption is neither correct or incorrect. What are the probabilities? I could be right or I could be wrong. I don't see a bottom. Losing 50 points or Handles in one month is quite crazy. btw, I consulted a Daily and Hourly chart for this.

To me it's about price action before indicators. The Gamma line (yellow) is a signal to buy or not buy. It will be more likely to move when the Gamma line is over 60ish. It may be a consolidation that will start, but with an earnings announcement due Monday AHours and an Expected Move of around $7-8.00 for Monday, I lean on the probabilities and say not yet.

Looking at the price top, you will see that Gamma was also at its bottom. What both of these Bottoming Gamma's mean is, you are correct, the momentum, whether up or down, is running out of steam. The thing we don't know is where it will stop. Just know that at the top, it's charged up and at the bottom it's exhausted. "That's what the Math Professor teaches."

I stay away from any stock that is riding the RSI on the bottom. The time or signal to buy is when the RSI crosses .2 and sell when the Blue RSI crosses below .8. Price action tells us if it's base building or consolidating. From 8-27 to 9-6 the RSILg made a quick trip up and down. Please notice how the price moved up then back from $150.00 to ~$170.00. Whew! now aren't you glad you asked this old man about this?

Addendum: after all of that blather, all I really wanted to emphasize is that beyond the nuances of this excellent indicator, the Gamma line should be considered a secondary signal. Some ORB day traders that have the RSILg on their charts have gamma as a label. I use Daily & Weekly and like to see it.

Hows that answer? My gamma line is around 20 and so is everyone reading this (still), God bless 'em. Time for bed.
 
Last edited:
MBF

MBF

Active member
2019 Donor
Warehouse
@markos yes I am having a hard time with this indicator. I understand better now about gamma being secondary. My dailys gamma is around 28. Hourly around 62 and rising and rsi rising also.
Thank you, always appreciative of your time and help.
 
P

pnguyen720

New member
Hey guys,
A little bit of a long shot but has anyone converted this indicator to Ninjascript? I'm relatively new to Thinkscript but have spent a lot of time inside NJT and would like to test this indicator.
Many thanks in advance.
 
markos

markos

Well-known member
VIP
@markos yes I am having a hard time with this indicator. I understand better now about gamma being secondary. My dailys gamma is around 28. Hourly around 62 and rising and rsi rising also.
Thank you, always appreciative of your time and help.
Hey @MBF your welcome. Have you read my tutorials on this? Find another stock and we can go thru it again. My preference is for ROKU and a Utility called NRG, but give me another. @Nick if you are available, please feel free to add to the conversation.
 
MBF

MBF

Active member
2019 Donor
Warehouse
@markos ROKU :love:
Looks great on the LaGuerre, chart looks good over 134. BUT ... on 7-26 it has the same little pattern on the LaG and that didn't pan out so well, different set up though.
How do you see it?
 
markos

markos

Well-known member
VIP
@markos ROKU :love:
Looks great on the LaGuerre, chart looks good over 134. BUT ... on 7-26 it has the same little pattern on the LaG and that didn't pan out so well, different set up though.
How do you see it?
@MBF I see the same thing on 1-25, 5-29 as well. Take everything in aggregate. This would have been sold if I listened to my second opinion TMO on 7-26. (I'm disobeying a cardinal rule in TA by having 2 similar indicators, don't be like me)
The RSIlg is telling me things are well. The stock got way ahead of itself and dropped from >160 to ~100 and now it's waiting for earnings. Notice that the ~40% drop didn't cut thru the 200 day EMA and it was given every opportunity to keep dropping but it made a lazy bottom instead. Going sideways at the 65D MA (No reason why I don't use a 50 Day) for over a week is a bonus for me. Also, it's not a sell signal until it drops below .8. Once that signal is given, I have to decide what I want to do. I am invested in this for my Grandkids, it will take a lot to break my faith. Hope that helps. Just my opinion. Markos
 
B

bdbald

New member
Could I please request some assistance on a scan please. It is for the RSI LaGuerre.
I just interested in finding those stocks that are at full scale overbought (1.0) and another scan for full scale oversold (0.00). I sure would appreciate it.

"B"
 
blakecmathis

blakecmathis

Active member
Warehouse


This is an RSIL study without the RSI just due to personal preference, so just FE. I would like to color code FE when it crosses above FEh, when it is between FEh and FEl, and when it crosses below FEl. Any help would be greatly appreciated!!
 
T

tomsk

Well-known member
VIP
Yes that is a standard method, all you got to do is to use AssignValueColor(). Here is a real simple example .
You can extract, adapt, modify and apply the method to your study

declare lower;
input n = 21;
def slope = MACD("fast length" = 12, "slow length" = 26, "macd length" = 9, "average type" = "SIMPLE");
def height = slope - slope[n];
plot data = Atan(height/n) * 180 / Double.Pi;
data.AssignValueColor(if data >=0 then Color.Cyan else Color.Yellow);
 
Last edited:
B

bdbald

New member
@markos Hello,,, At the top of the thread in that chart of BYND, the RSI is light blue line is showing full scale around 8/6 to 8/14. Is there a way to scan for stocks to find these that are full scale flat like this? thank you
 
markos

markos

Well-known member
VIP
@markos Hello,,, At the top of the thread in that chart of BYND, the RSI is light blue line is showing full scale around 8/6 to 8/14. Is there a way to scan for stocks to find these that are full scale flat like this? thank you
@bdbald sure, there are a number of them posted throughout these threads. Actually, the buy trigger is when the RSI Laguerre is crossing above the .20. Then you make sure that the price action is confirming what the indicator is saying. There are a good amount of notes on this in the tutorials section. Good luck!
 
B

bdbald

New member
@markos Thanks for the kind reply.. I was taught what you had mentioned about crossing the 0.20. What I was hoping to find were those stocks at full scale 1.00 and 0.00. Then I wanted to apply a different proprietary indicator from there. I went to Tutorials, Questions, and Indicators Sections for RSI LaGuerre and couldn't exactly see what I was shooting for. If you could lend a hand I sure would appreciate it. I am nowhere in the same league as you guys tech-wise.... "B"
 
horserider

horserider

Well-known member
VIP
@bdbald Maybe something like this.

# Scan for RSI in Laguerre Time With Fractal Energy
# Mobius
# V02.07.2014
# V03.06.15.2016
#Scan
#Inputs:
input nFE = 8;
# Variables:
def o;
def h;
def l;
def c;
def CU1;
def CU2;
def CU;
def CD1;
def CD2;
def CD;
def L0;
def L1;
def L2;
def L3;
def RSI;
# Calculations
o = (open + close[1]) / 2;
h = Max(high, close[1]);
l = Min(low, close[1]);
c = (o + h + l + close) / 4;
def gamma = Log(Sum((Max(high, close[1]) - Min(low, close[1])), nFE) /
(Highest(high, nFE) - Lowest(low, nFE)))
/ Log(nFE);
L0 = (1 – gamma) * c + gamma * L0[1];
L1 = -gamma * L0 + L0[1] + gamma * L1[1];
L2 = -gamma * L1 + L1[1] + gamma * L2[1];
L3 = -gamma * L2 + L2[1] + gamma * L3[1];
if L0 >= L1
then {
CU1 = L0 - L1;
CD1 = 0;
} else {
CD1 = L1 - L0;
CU1 = 0;
}
if L1 >= L2
then {
CU2 = CU1 + L1 - L2;
CD2 = CD1;
} else {
CD2 = CD1 + L2 - L1;
CU2 = CU1;
}
if L2 >= L3
then {
CU = CU2 + L2 - L3;
CD = CD2;
} else {
CU = CU2;
CD = CD2 + L3 - L2;
}
RSI = if CU + CD <> 0 then CU / (CU + CD) else 0;
# Note Comment-Out (#) whichever scan not being used.
# Long Scan
plot Long = RSI = 0 ;
# Short Scan
#plot Short = RSI = 1 ;
 
B

bdbald

New member
@horserider Wow! Can't thank you enough! Hope it gets me there...
 
B

bdbald

New member
@horserider Well,, When I put the code in, I had a red error and I guess it was prompting me for the double equal signs on the two "plot Long" and "plot Short" lines at the very end... I did that and it cleared. Going through the returns, a very few a < 0.25, but not 0.00, but the clear majority are just fine at 0.00 so great! One thing is weird, as a saved search,,, the TOS watchlist is not automatically/dynamically loading the results... I have heard this happens from time to time. Thanks so much again, sir.. I think you got me there..
 
T

tomsk

Well-known member
VIP
@bdbald the plot condition that @horserider posted looked different, so I fixed it up. This version scans for last 4 bars above 0.9 or below 0.1. Make sure you select ONLY one plot and comment the other out as appropriate.

I just ran a scan against the S&P 500 using this code and it returned 96 hits. Feel free to adjust your scan condition as you see fit

# RSI in Laguerre Time With Fractal Energy Scan
# Mobius
# V02.07.2014
# V03.06.15.2016
#Scan
#Inputs:
input nFE = 8;
# Variables:
def o;
def h;
def l;
def c;
def CU1;
def CU2;
def CU;
def CD1;
def CD2;
def CD;
def L0;
def L1;
def L2;
def L3;
def RSI;
# Calculations
o = (open + close[1]) / 2;
h = Max(high, close[1]);
l = Min(low, close[1]);
c = (o + h + l + close) / 4;
def gamma = Log(Sum((Max(high, close[1]) - Min(low, close[1])), nFE) /
(Highest(high, nFE) - Lowest(low, nFE)))
/ Log(nFE);
L0 = (1 – gamma) * c + gamma * L0[1];
L1 = -gamma * L0 + L0[1] + gamma * L1[1];
L2 = -gamma * L1 + L1[1] + gamma * L2[1];
L3 = -gamma * L2 + L2[1] + gamma * L3[1];
if L0 >= L1
then {
CU1 = L0 - L1;
CD1 = 0;
} else {
CD1 = L1 - L0;
CU1 = 0;
}
if L1 >= L2
then {
CU2 = CU1 + L1 - L2;
CD2 = CD1;
} else {
CD2 = CD1 + L2 - L1;
CU2 = CU1;
}
if L2 >= L3
then {
CU = CU2 + L2 - L3;
CD = CD2;
} else {
CU = CU2;
CD = CD2 + L3 - L2;
}
RSI = if CU + CD <> 0 then CU / (CU + CD) else 0;
# Note Comment-Out (#) whichever scan not being used.
# Long Scan
plot Long = sum(RSI > .9, 4) >= 4;
# Short Scan
#plot Short = sum(RSI < .1, 4) >= 4;

# END SCAN
 
Last edited:
B

bdbald

New member
Thank you @tomsk ! Yes,, I was deselecting one and using the other... Looking forward to trying yours also..
 
B

bdbald

New member
@horserider and @tomsk I tried to admit my "special ed" status in this whole department.... In horserider's scan, I just noticed his "nfe" was different than mine, I altered it, and I get perfect results.... Duh to me of course,,, sorry for the confusion and thanks again... What I am doing is using this as a starting point, then applying a Ready-Aim-Fire or FW_Fisher Transformer to it.... I've seen a lot of potential with it.. Thanks again..... "B"
 
horserider

horserider

Well-known member
VIP
@bdbald Since you us FW FT I will suggest RSI Inverse Fischer Transform.
The FW FT is too confusing for me.
 

Top