Chandra's Indicators (with Alerts)

Quoting achandrasekaran99

Disliked

Ohka and KingSizePips provided TS code for the regular Laguerre in the main forum. I will try to modify to get the MA= 1 - 3 (I don't know how to make general loop in TS). Since I have never seen TS, I am not sure it will work. If it generates any error or the values don't match the Laguerre-ACS1, Ohka can correct it.

Hi Ohka, if you read this, please check it and correct if necessary. Thanks.

Hope it works. Let us know.
chandra

{Start of TS-Laguerre-ACS1 code}
inputs: gamma(0.5), MA(2) ;
Vars: L0(0) ,
L1(0) ,
L2(0) ,
L3(0) ,
CU(0) ,
CD(0) ,
dummy(0),
LaguerreRSI(0) ;

L0 = (1 - gamma) * close + 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] ;

CU = 0 ;
CD = 0 ;
if L0 >= L1 then CU = L0 - L1 Else CD = L1 - L0 ;
if L1 >= L2 then CU = CU + L1 - L2 Else CD = CD + L2 - L1 ;
if L2 >= L3 then CU = CU + L2 - L3 Else CD = CD + L3 - L2 ;

if CU + CD <> 0 then dummy = CU / (CU + CD) ;
if MA < 2 then LaguerreRSI = dummy ;
if MA = 2 then LaguerreRSI = (dummy + dummy[1])/2 ;
if MA > 2 then LaguerreRSI = dummy + dummy[1] + dummy[2])/3 ;

Plot1(LaguerreRSI, "LaguerreRSI") ;
Plot2(.85) ;
Plot3(.15) ;
{End of TS-Laguerre-ACS1 code}

Ignored

Quoting achandrasekaran99

Disliked

Hi Ohka,

I worked on Fisher Transform on the price itself in another thread (about four months back). First we had a re-painter that looked too good on visual backtesting. Then I fixed the code and even made alert similar to LISPA (actually LISPA is an improved version of that alert system). After a lot of work, we found that it was adding too much delay and abandoned it. I don't know how it will work on the LagRSI. But I think it will add similar delay here also. Just the Lag-MA, Stoch, MT4-LSAR should be enough I think.

BTW, If you want to plot two Lag-MAs, just add them twice with two different MAs on the same window (you can use 1 or 0 as MA value to get regular Lag). Then you can even see moving average crossover. I checked with 2-4 MA and it looked OK.

chandra

PS: I use ACS for Laguerre (and many other programs as a kinda abbreviation of my name). It is confusing with Imran's ASC indicator it looks! At least for me! When you write ASC, I am thinking of that Imrans ASC indicator. No problem, just confusing.

Ignored

Quoting eagle4x

Disliked

Al: There is a HUGE difference between the regular Laguerre and Chandra's indicator (btw - great job Chandra!). If you don't change the MA setting from the default of "2", then it does look the same as regular Laguerre. You need to increase the MA setting to reduce the whipsaws. Also, increase gamma to improve signal quality. Tweak the settings based on whatever timeframe you are using. I was using the Awesome indicator to reduce whipsaws of the regular Laguerre, but Chandra's MA addition to smooth it out is so much better that it doesn't need a confirming indicator.

Ignored

Quoting Ohka

Disliked

Hi Chandra & others,

the attached bitmap shows my current settings for your Laguerre-ACS indicator. I've found a combination that makes me comfortable enough to remove the normal Laguerre indicators.

In keeping with Imran's system, I still use two Laguerre-ACS'ses, with the following details:

MA: 3 on both Lag-ACS
Gamma 1 (blue): 0.5
Gamma 2 (red): 0.67
threshold levels 0.9, 0.5 & 0.1

I use BollingerBands, the 200 & 60 EMA and StochHist (all per Imran's directions) to be able to put signals by Lag-ACS into perspective.

take care
Ohka

Ignored

Quoting Sailingalbob

Disliked

Thanks Chandra and Ohka,

Your explanations and examples are extremely useful to understand.

AL

Ignored

Quoting Ohka

Disliked

Hi Chandra & others,

the attached bitmap shows my current settings for your Laguerre-ACS indicator. I've found a combination that makes me comfortable enough to remove the normal Laguerre indicators.

In keeping with Imran's system, I still use two Laguerre-ACS'ses, with the following details:

MA: 3 on both Lag-ACS
Gamma 1 (blue): 0.5
Gamma 2 (red): 0.67
threshold levels 0.9, 0.5 & 0.1

I use BollingerBands, the 200 & 60 EMA and StochHist (all per Imran's directions) to be able to put signals generated by Lag-ACS into perspective.

take care
Ohka

Ignored

Quoting driftwood

Disliked

hello chandra and ohka - thx for all work u do here Can i tell is this indi the one on 1st post or been change now?

Ignored

Quoting eagle4x

Disliked

Chandra:
I suggest modifying the indicator so that the type of MA can be specified - exponential, simple, etc., (the same options as MA indicator in MT). Would you mind doing that and posting it in this thread?

Ignored

Quoting achandrasekaran99

Disliked

Hi eagle,

Thank you for this suggestion. A good idea. So far, I have been calculating this type of MA by simple addition/divison. May be it is time for me to use the sophisticated method built-in MT4. Only that can do all these options. I will include it in the next version. Thanks.

chandra

Ignored

Quoting Ohka

Disliked

Let's try and remain realistic about what we actually can achieve... All variations of Chandra's Laguerre we have been discussion so far will signal an entry exactly when Imran's original system does so - or perhaps only one single bar later at the most.
I feel it would be much more worthwhile to think about adding filters to weed out the fake signals - much more so than tweaking this already fine indicator to two decimals behind the comma.
Don't get me wrong, new ideas should be tried. However, that being said, the animal is really good already and any improvement from here on can only be a marginal one and hence that effort would be a poor deployement of resources (80/20-rule, remember?).

Let's throw around new filtering ideas, that define/limit the context within which Chandra's Laguerre should be used....

Just my 0.02€...

Ohka

Ignored

Quoting achandrasekaran99

Disliked

Hi Ohka,

Thank you for your detailed analysis of the variations to MA method. I was thinking only as a programer and did not think about its real effect on our numbers. Now, I also agree with you. When we use small MA such as 3 bars, the variations will not have much effect. Only when longer MAs are used, these will have significant effect. So, may be that is not important for the numbers we are going to use. I hope eagle agrees with this. Then we can keep it as such and move on to next part.

BTW, the filters are not part of this Indicator. They will come in LISPA2, whenever the filters are ready (Imran has suggested some similar to the ones in the PDF document). I will have to start work on that.

Any filtering ideas for LISPA2 are welcome. (Though I will pass them first to Imran.)

Thank you both for the thoughts you put into this. The more ideas we pass through, the better the method/system will be at the end (even if not all the ideas are used).

chandra

Ignored

Quoting eagle4x

Disliked

Chandra:
After trying the laguerre_ACS1 and regular laguerre with different settings, I found that the regular laguerre works better for me. Attached is a gbpusd 1hr chart with the regular laguerre and an indicator called awesome to be used for filtering. I set gamma on laguerre to .86.

To filter with awesome, use "+" and/or green for long if laguerre crosses above .15, and for short "-" and/or red if laguerre crosses below .90. The default MA for awesome is 34 SMA, however, I changed the code to 17 SMA to make it more responsive. Maybe you have some ideas for tweaking awesome for improving the filtering.

Ignored

Quoting euro_dude

Disliked

Hi Chandra:

Can you add the Bid Price to the TextSignal (i.e. SELL @ 1.3323) of your SAS2 indicator?

I don't find the inserted TIME (i.e. SELL [21:10]) particularly useful since the arrow location essentially does the same.

Regards,

Ignored

Quoting achandrasekaran99

Disliked

Hi eagle,

Thank you for the input. The gamma you are using looks like very big number. When you are looking that big, surely MAs will not help much. It is very useful in reducing lot of fakes for the commonly used gamma of 0.55, especially in the lower timeframes.

In the Imran's main thread, someone mentioned that gamma=0.35 was working good in hourly chart. May be that was for different pair and different holding period.

You can use my Indicator for regular Lag also (just use MA=1 or less). It will atleast put the gamma value in the chart so that everyone can see what settings were used on that chart (for this indicator atleast).

Currently I am working on LISPA2 (with Imran). But I will keep in mind about the Awesome Oscillator and check it later.

Thanks.
chandra

Ignored

Quoting eagle4x

Disliked

Chandra:
See attached 1hr chart with gamma setting of .35. Notice the multiple whipsaws. Compare to the previous chart I posted which has very few whipsaws - so I do not understand why everyone on this thread is saying to lower gamma. If I am overlooking something, fill me in.

Ignored