@Mike How do I use the StateFIFOQueue to track my paintbars performance (let's say, the winrate of a paintbar on a particular security?)

[Thebattlefront]

I am aware of using the GetHighLow function to track the yearly high and yearly low at a particular bar using the StateFifoQueue, aka:
 

HighLow.Add(High);

HighLow.Add(Low);

double High52, Low52;

HighLow.GetHighLow(out High52, out Low52);
 

I was wondering, though, how do I create a variable that tracks, let's say, the average percent gain per bar of a security
 (m_dGainPercentAveragePerBarPerSecurity), or maybe something like the Average winrate of a paintbar (m_dAverageWinrate),
 or some other arbitrarily defined variable? You mentioned in another thread i could use Statefifoqueue to take averages over
a smaller space rather than averages over the entire space, and I assume this relates to calculation of these values in some way.

Edited October 17, 2022 by Thebattlefront

[Mike Medved]

It's the same way. You define a new StateFIFOQueue(n) - where n is the period. You use the state function to save state/restore state for that queue.

Every maincalc you .Add whatever it is to the queue. The Queue's Average property will hold the average. It also has StDev, WeightedAverage etc...

 

[Thebattlefront]

Okay, I think I understand. I also found the medved help article on this so I probably should have looked there first. 

Thanks Mike

[Thebattlefront]

One more thing - what's the syntax difference between using Q.WeightedAverage and Q.Average?

Since the fifoqueue Q only holds a single set of real values (let's say, for example, Q = {1, 2, 3, 4, 5}), how do I apply a weight to Q?

The reason I ask is because, it would appear that I can use Q.WeightedAverage or Q.Average on the same line and it returns no errors. But where are the weights coming from exactly in the program when I set a variable equal to Q.WeightedAverage? Does it just apply a weight of 1 to each value if I don't specify a new set P that contains the weights for Q?

Edited October 20, 2022 by Thebattlefront

[Mike Medved]

The weight is a 1-based position in the queue from start to end. So for your example it is (1*1+2*2+3*3+4*4+5*5)/(1+2+3+4+5)