How can I make Keltner RSI?

[hybinubt]

The Keltner RSI uses the simple moving average, whereas Wilder's RSI, as provided by NinjaTrader, employs Wilder's moving average (similar to EMA).

I attempted to create a Keltner RSI indicator using the NinjaScript editor, but I encountered repeated challenges.

Below is the logic I devised for the Keltner RSI. Instead of using the closing price (C), I utilized the Typical Price (T), defined as:
T = (C + H + L) / 3

The formula for Keltner RSI is as follows:
RSI(Period) = 100 * [Sum(if (T - T(1) > 0, T - T(1), 0), Period)] / [Sum(if (T - T(1) > 0, T - T(1), T(1) - T), Period)]


Explanation:

1. Upward Changes:
   Sum(if (T - T(1) > 0, T - T(1), 0), Period)
   • If the current Typical Price (T) is higher than the previous bar’s Typical Price (T(1)), add the difference (T - T(1)).
   • Otherwise, add 0. This represents the sum of upward changes over the period.
2. Upward + Downward Changes:
   Sum(if (T - T(1) > 0, T - T(1), T(1) - T), Period)
   • If the current Typical Price (T) is higher than the previous bar’s Typical Price (T(1)), add the difference (T - T(1)).
   • Otherwise, add the absolute difference in reverse order (T(1) - T).
   • This results in the total of upward and downward changes over the period.
3. RSI Calculation (Upward/Total Changes):
   Divide the sum of upward changes (Section 1) by the total changes (Section 2), and multiply the result by 100 to convert it into a percentage.

Is there any way to make Ninjascript indicator that follows above logic?

[NinjaTrader_ChelseaB]

Hello hybinubt,

Below is a link to a support article with helpful resources on getting started with C# and NinjaScript.

https://support.ninjatrader.com/s/article/Developer-Guide-Getting-Started-with-NinjaScript

What challenges are you having?
What have you coded so far?

This doesn't appear to be NinjaScript or C#. Is this written in a different language?

Where you have:
T = (C + H + L) / 3

I think this translates to:
MyTypicalSeries[0] = (Close[0] + High[0] + Low[0]) / 3;

Where you have:
RSI(Period) = 100 * [Sum(if (T - T(1) > 0, T - T(1), 0), Period)] / [Sum(if (T - T(1) > 0, T - T(1), T(1) - T), Period)]

I think this translates to:
Value[0] = 100 * new int[] { Math.Max(0, MyTypicalSeries[0] - MyTypicalSeries[1]), Period }.Sum() / new int[] { Math.Max(0, MyTypicalSeries[1] - MyTypicalSeries[0]), Period }.Sum();