Asymmetrical RSI (ARSI) - Formula
Author: tradecom
Creation Date: 9/20/2019 8:37 PM
profile picture

tradecom

#1
Hello,

I wan't to calculate the Asymmetrical RSI (TASC 2008-10) at Excel, but I can't find the right formula.
I calculated the RSI, which was no problem, but with the formula for ARSI I can't get the same value as the strategy calculated in WealthLab.

Thanks!
profile picture

Eugene

#2
Hi,

The TASCIndicators library's code is open source. You can review the ARSI formula here:

Home - TASCIndicators
profile picture

tradecom

#3
Hello Eugene,

thanks for the information. I calculated the ARSI like you descriped it.

There is just one thing I don't understand:

I have an ARSI period of 9 days. In this 9 days I have 6 Upmoves and 3 Downmoves, so I have at upcount 6 and 3 at downcount. But the endresult is not the same as calculated by wealth lab. If i take 4 for upcount and 5 for downcount I get the same result. Is there a special rule for counting the upmoves and downmoves?

Thank you!

Kind regards

Ivana
profile picture

Eugene

#4
Hi Ivana,

Could you clarify your problem here in more detail? In the first post you talked about duplicating the WL formula in Excel (which is not supported by us). If you have a concern about our implementation for WL which you can illustrate, please feel free to provide an example.
profile picture

tradecom

#5
Hello Eugene,

thanks for your response.

In your ARSI strategy description you write:

CODE:
Please log in to see this code.


So Upcount are all positive Performances of a period and Downcount is Period minus Upcount. So i have, as explained above, 6 values for upcount and 3 for downcount, but Wealth Lab is calculating with 4 values for upcount and 5 for downcount in the same case.

Is there another special description for calculating the UpCount?

Thank you!

Kind regards
profile picture

Eugene

#6
Actually, this line of code does not calculate an "up count":
CODE:
Please log in to see this code.

Here's the key, a couple of lines up:
CODE:
Please log in to see this code.

The CumUp function with period of 1 (as here) returns the number of consecutive bars where some DataSeries was greater than their value a number of bars ago (here: 1). This should explain the difference you're getting.
profile picture

tradecom

#7
Hello Eugene,

thank you.

Your explanation is exactly what I did. I have a period of 9 days. In this 9 days I have 6 Upcount and 3 Downcount. But Wealth Lab is taking 4 for Upcount and 5 for Downcount. In periods where I have no Upcount over a long period, Wealth lab is calculating with for example 2 or 4 upcounts. Are there more rules for getting Upcount?

Thanks!

Ivana
profile picture

Eugene

#8
Ivana,

Sorry but I'm at a loss as to what you did in Excel and you probably agree with me on that it doesn't make much sense for us to discuss your implementation as this is not a support matter. I have a feeling that we're going round in cycles. I'm pretty sure that the Wealth-Lab numbers are correct but at the risk of repeating, feel free to point me at the issue if you're seeing it in Wealth-Lab. For convenience you might want to plot the CumUp indicator to visualize how a consecutive up/down count is calculated. Attach a screenshot (in PNG format) if required. Thanks.
This website uses cookies to improve your experience. We'll assume you're ok with that, but you can opt-out if you wish (Read more).