What Am I Missing with My Slope Calculation Code ?

[Aston01 10]

I seem to be having a little bit of an issue with my slope calculation and was hoping someone could double check that I had written it properly.

 

The issue is basically that even with the slope degree input adjusted to extremes, some of these flat spots still trigger differently. If working properly the small transition in the middle should have stayed magenta as opposed to going green for 2-3 bars and back to magenta. Am I missing something in my code ?

 

 

 

Here is a snippet of the EasyLanguage code in question.

 

Inputs: 
SlopeAv(5); 

Vars; 
Slope(0); 

Avg = average( Close, length) ; 
Slope =(TLSlopeEasy(Avg,1,SlopeAv)); 

if colorMode = 1  
then setplotcolor[1](1, iff( Avg <Avg [1] and (Slope < .01 ), magenta, green ) ); 
plot1 (Avg , "Avg ");

 

 

Below is a snippet from the diagnostics file (area highlighted in blue is where the color shift occurs). Am I missing something ...seems like just the type of scenario I was trying to smooth. The answer could be staring me in the face and I might not be seeing it.

 

------------------------------------------------------------------------------

1111214 1419Avg 3.48Avg[1] 3.51|slope -0.04|condition FALSE

1111214 1419Avg 3.45Avg[1] 3.48|slope -0.03|condition FALSE

1111214 1420Avg 3.42Avg[1] 3.45|slope -0.03|condition FALSE

1111214 1420Avg 3.40Avg[1] 3.42|slope -0.03|condition FALSE

1111214 1420Avg 3.38Avg[1] 3.40|slope -0.03|condition FALSE

1111214 1420Avg 3.37Avg[1] 3.38|slope -0.02|condition FALSE

1111214 1420Avg 3.36Avg[1] 3.37|slope -0.02|condition FALSE

1111214 1420Avg 3.35Avg[1] 3.36|slope -0.02|condition FALSE

1111214 1421Avg 3.34Avg[1] 3.35|slope -0.01|condition FALSE

1111214 1421Avg 3.34Avg[1] 3.34|slope -0.01|condition FALSE

1111214 1423Avg 3.35Avg[1] 3.34|slope -0.01|condition FALSE

1111214 1424Avg 3.35Avg[1] 3.35|slope -0.00|condition FALSE

1111214 1426Avg 3.35Avg[1] 3.35|slope 0.00| condition FALSE

1111214 1429Avg 3.36Avg[1] 3.35|slope 0.00| condition FALSE

1111214 1429Avg 3.36Avg[1] 3.36|slope 0.00| condition FALSE

1111214 1429Avg 3.36Avg[1] 3.36|slope 0.00| condition FALSE

1111214 1430Avg 3.36Avg[1] 3.36|slope 0.00| condition FALSE

1111214 1430Avg 3.36Avg[1] 3.36|slope 0.00| condition FALSE

1111214 1430Avg 3.36Avg[1] 3.36|slope 0.00| condition FALSE

1111214 1431Avg 3.35Avg[1] 3.36|slope -0.00|condition FALSE

 

Any help would be much appreciated

[onesmith 10]

The code you posted is not the code that produced the chart you posted. When asking for help it is best if your chart includes the symbol name in the picture. The symbol should be set to exchange time. This enables calibrating a test case to the aproximately 14:30 time (on this unknown symbol) using some code other than the code you posted.

[Tradewinds 12]

the middle should have stayed magenta as opposed to going green for 2-3 bars and back to magenta. Am I missing something in my code ?

 

if colorMode = 1

then setplotcolor[1](1, iff( Avg <Avg [1] and (Slope < .01 ), magenta, green ) );

plot1 (Avg , "Avg ");

 

The color is controlled by 'Avg' being up or down since the last bar. Avg <Avg [1]. If you wanted it delayed, wouldn't you need to check multiple bars?

 

To change color down it would need to pass this test:

 

Avg <Avg [1] and Avg[1] < Avg [2] and Avg[2] < Avg [3]

 

otherwise stay the same color, or display a neutral color.

 

var: ChangeDown(False), ChangeUp(False);

 

ChangeDown = Avg <Avg [1] and Avg[1] < Avg [2] and Avg[2] < Avg [3];

ChangeUp = Avg >Avg [1] and Avg[1] > Avg [2] and Avg[2] > Avg [3];

 

if ChangeDown then

setplotcolor[1](1, magenta )

else if ChangeUp then

setplotcolor[1](1, Green );

[Do Or Die 10]

LOL, just curious as onesmith, how does the software actually calculates slope.

[onesmith 10]

The formula for slope is slope=(endprice-startprice)/length; A threshold can be optimized to filter out undesirable slope levels.

input: threshold(.1);
var: avg(0), slope(0);

avg=average(c,10); 
slope=(avg-avg[10])/10;

if slope > threshold then setplotcolor(1,green) 
else if slope < -threshold then setplotcolor(1,red);

plot1(avg,"avg");

[Do Or Die 10]

Thanks for confirming, that is the classical definition. Many people use the regression line slope also.

[tommaso 10]

Thanks for confirming, that is the classical definition. Many people use the regression line slope also.

 

another effective (and more robust) way to determine direction (actually a form of "slope") can be the SDX - signed direction index

 

SDX

 

Tom

[onesmith 10]

another effective (and more robust) way to determine direction (actually a form of "slope") can be the SDX - signed direction index

 

SDX

 

Tom

 

Tom, I'm trying to translate sdx. Please post source code. From the formula, I understand S=Speed but I don't understand the line below.

 

S(i) = P(i) - P(i-1) i= 1, ... , k-1

 

If i represents an index from 1 to 10 what does k represent? ... a 1 point increment in grid? Can you help me visualize k in the code below?

 

for i=1 to 10 begin
   s(i)=p(i)-p(i-1);
end;

[tommaso 10]

Hi onesmith

 

thanks for asking.

 

Since you are interested in a recent value of the "direction", you limit your computations to a recent number of price variations ("moves" from a gridline to another). The (k-1) would represent how many latest distinct price changes you are considering to assess the recent "direction".

 

When you use equispaced gridlines (and do not use weighting), you can just simplify as follows (which is immediately intuitive to visualize):

             Number of "UpMoves" - Number of "DownMoves"
sdx =     --------------------------------------------------------------------   *  100
                              Number of all "Moves"  

 

intending here "UpMoves", "DownMoves", "Moves" as the distinct price "moves" from a gridline to the adjacent one.

 

[ The distance between gridlines depends on your "timeframe" (similar to when you measure time, you may be interested in microseconds variations, or you may just be intested in minutes, depending on what you are doing).

 

(I often use a distance between gridlines equal to 0.1% of the price.) ]

 

Let me know if something unclear here.

 

 

Tom

 

 

Tom, I'm trying to translate sdx.

[...]

If i represents an index from 1 to 10 what does k represent? ... a 1 point increment in grid? Can you help me visualize k in the code below?

 

for i=1 to 10 begin
   s(i)=p(i)-p(i-1);
end;

Edited December 22, 2011 by tommaso