RSI (sma) code, does it look correct?

[Johan Strandberg]

Hi, am programming in C# and wants to use my code to backtest RSI settings that I then wish to program in MQL. 
DOES THE CALCULATION LOOK CORRECT BELOW? I'm using SMA-method for the calculation.

In the example below I will have a setting of RSI 5

1) Calculate ChangeUp and ChangeDown for each bar by looking at current bar closeprice vs diff of previous bar closeprice.

2) For current bar RSI I add up all ChangeDown(and same for ChangeUp) for the last 4 bars + current (yes, I won't know RSI until current bar closes) / N bars. This gives me an AverageUp and AverageDown for the current bar. 

3) Calculate RS by taking AverageUp/AverageDown. 

Then RSI for the CURRENT bar is = (100 - (100 / (1 + RS)));

ANY COMMENTS ON MY CALCULATIONS CORRECTNESS?

[William Roeder]

Johan Strandberg: 2) For current bar RSI I add up all ChangeDown(and same for ChangeUp) for the last 4 bars + current (yes, I won't know RSI until current bar closes) / N bars.

That isn't the definition. It's not a SMA. The ChangeUp is a (exponential) moving average.

Average Gain = [(previous Average Gain) x 13 + current Gain] / 14.

           Relative Strength Index (RSI) [ChartSchool]

That isn't the definition. Look at the MT4 code

diff=close[i]-close[i-1];
ExtPosBuffer[i]=(ExtPosBuffer[i-1]*(InpRSIPeriod-1)+(diff>0.0?diff:0.0))/InpRSIPeriod;
ExtNegBuffer[i]=(ExtNegBuffer[i-1]*(InpRSIPeriod-1)+(diff<0.0?-diff:0.0))/InpRSIPeriod;
if(ExtNegBuffer[i]!=0.0)
   ExtRSIBuffer[i]=100.0-100.0/(1+ExtPosBuffer[i]/ExtNegBuffer[i]);

[Fernando Carreiro]

Johan Strandberg:

Hi, am programming in C# and wants to use my code to backtest RSI settings that I then wish to program in MQL. 
DOES THE CALCULATION LOOK CORRECT BELOW? I'm using SMA-method for the calculation.

In the example below I will have a setting of RSI 5

1) Calculate ChangeUp and ChangeDown for each bar by looking at current bar closeprice vs diff of previous bar closeprice.

2) For current bar RSI I add up all ChangeDown(and same for ChangeUp) for the last 4 bars + current (yes, I won't know RSI until current bar closes) / N bars. This gives me an AverageUp and AverageDown for the current bar. 

3) Calculate RS by taking AverageUp/AverageDown. 

Then RSI for the CURRENT bar is = (100 - (100 / (1 + RS)));

ANY COMMENTS ON MY CALCULATIONS CORRECTNESS?

Please stop double posting! You already had open thread on this subject at: https://www.mql5.com/en/forum/214849

On that thread, we had already given answers! You should have continued your comments there.

[Johan Strandberg]

whroeder1:
That isn't the definition. It's not a SMA. The ChangeUp is a (exponential) moving average.

           Relative Strength Index (RSI) [ChartSchool]

That isn't the definition. Look at the MT4 code

Thanks Whroeder1, the coin dropped now I think. Really appreciate your help. 

So I've been using this calculation and choose the SMA-calculation of AverageUp and AverageDown whereas the MT4 indicator uses an EMA-calculation of AverageUp and AverageDown. 

Thanks

[Johan Strandberg]

Fernando Carreiro:

Please stop double posting! You already had open thread on this subject at: https://www.mql5.com/en/forum/214849

On that thread, we had already given answers! You should have continued your comments there.

Your answers were not answers nor in this or the other thread. 

[Mladen Rakic]

Johan Strandberg:

Thanks Whroeder1, the coin dropped now I think. Really appreciate your help. 

So I've been using this calculation and choose the SMA-calculation of AverageUp and AverageDown whereas the MT4 indicator uses an EMA-calculation of AverageUp and AverageDown. 

Thanks

In that case you are using what is known as Cutler's RSI

See here for more information : https://en.wikipedia.org/wiki/Relative_strength_index

---

PS: original Welles Wilder RSI uses what is sometimes known as Wilders EMA (like bellow) not ema like in the version you are talking about :

EMA formula = price today * K + EMA yesterday * (1-K) where K = 2 / (N+1)

Wilder EMA formula = price today * K + EMA yesterday (1-K) where K =1/N

Where N = the number of periods.

Even though they look different, Wilders EMA and SMMA are producing same results, so you can use SMMA too

[Fernando Carreiro]

Mladen Rakic:

In that case you are using what is known as Cutler's RSI

See here for more information : https://en.wikipedia.org/wiki/Relative_strength_index

---

PS: original Welles Wilder RSI uses what is sometimes known as Wilders EMA (like bellow) not ema like in the version you are talking about :

Even though they look different, Wilders EMA and SMMA are producing same results, so you can use SMMA too

Be careful of this OP! I provided similar information to him (see below) on his other thread and I was insulted by him for it, with the following (which he has since removed):

Thank you mr. Nerd ... Wow, you know how to google ... did it ever struck that I also know how to google and asked a specific question which you didn't answer nor provided any clue on how to solve ... Crappy answer my friend, crappy answer indeed.

Forum on trading, automated trading systems and testing trading strategies

Understanding RSI calculation

Fernando Carreiro, 2017.09.01 13:35

Have a look at WikiPedia's page on RSI (Relative Strength Indicator) which offers you a complete explanation of how it is calculated.

You can also consult the original source, which is the 1978 book by J. Welles Wilder, New Concepts in Technical Trading Systems, that has an example step-by-step breakdown on how it is calculated.

Forum on trading, automated trading systems and testing trading strategies

Understanding RSI calculation

Johan Strandberg, 2017.09.08 14:28

You do know that RSI AveageUp and AverageDown can be calculated using SMA, EMA and Wilders Smooting right?

Just asking as your answer did not contain anything of value to my question that google had not already provided.....

[Mladen Rakic]

Fernando Carreiro:

Be careful of this OP! I provided similar information to him (see below) on his other thread and I was insulted by him for it, with the following (which he has since removed):

Don't worry :) I am not known for being too quiet :)

[Fernando Carreiro]

Mladen Rakic: Don't worry :) I am not known for being to quiet :)

OK!

[Johan Strandberg]

Fernando Carreiro:

OK!

Forgive me for being rude but I just wasn't interested in a book to read, I was interested in somebody who had knowledge who could tell me if my code looks correctly based on the information I've already read and now posted. 

Feel free to move on and help other people instead of me and also stop spamming my thread. 

Kind Regards

[Johan Strandberg]

Mladen Rakic:

In that case you are using what is known as Cutler's RSI

See here for more information : https://en.wikipedia.org/wiki/Relative_strength_index

---

PS: original Welles Wilder RSI uses what is sometimes known as Wilders EMA (like bellow) not ema like in the version you are talking about :

Even though they look different, Wilders EMA and SMMA are producing same results, so you can use SMMA too

Thanks Mladen, 

Interesting that they produce the same result (and that it's called Cutlers when using SMA). Would you know if my SMA-code looks correct? I think it's okey but would be great if I could get it confirmed. 

I've followed macrooptions link and then I got unsure when looking at a regular RSI in MT4, in retrospect I now understand with the help of some previous answers that the MT4 RSI is calculated with a EMA which is why my own code based on SMA differs :)