The adx calculation formula is wrong.

[vnvdev]

I use MT5 and Talib to calculate ADX, however, produce 2 completely different results!

[mrjbq7]

Can you provide an example of the data and the two outputs, showing which one you think should be correct?On Jan 24, 2023, at 5:27 PM, Vũ Nhật Vượng @.***> wrote: I use MT5 and Talib to calculate ADX, however, produce 2 completely different results!

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you are subscribed to this thread.Message ID: @.***>

[vnvdev]

https://github.com/vnvdev/adx-calculate-same-mt5/blob/main/adx.py Here is the code that compares my 2 outputs and the adx on talib. For talib, adx is only true on adx 14, the rest of the lengths are wrong

[vnvdev]

The reason I want to use talib is because it is faster than my code, but the results with adx7, adx21, adx other than 14 all give wrong results compared to mt5

[vnvdev]

As for my code all adx are correct with mt5

[vnvdev]

Hope you fix it soon

[trufanov-nok]

Have you tried talib.set_compatibility(1) before ADX call?

[vnvdev]

Have you tried talib.set_compatibility(1) before ADX call?

What is this?

[trufanov-nok]

https://stackoverflow.com/questions/74896623/python-ema-calculation-using-talib-and-pandas-ewn-different-from-tradingview

Have you tried talib.set_compatibility(1) before ADX call?

What is this?

[vnvdev]

https://stackoverflow.com/questions/74896623/python-ema-calculation-using-talib-and-pandas-ewn-different-from-tradingview

Have you tried talib.set_compatibility(1) before ADX call?

What is this?

This not work with adx

[vnvdev]

Im fixed it with my Cython code, it very fast same talib. Here is my code https://github.com/vnvdev/adx_faster

[pablobrunetti]

@vnvdev, is this code work with Plus Directional Indicator(PLUS_DI ) and Minus Directional Indicator(MINUS_DI)? I had the same problem.

[vnvdev]

The code only calculates the adx index, the other indicators do not