Average-of-multiple-TimeSeries-with-data-gaps-and-limit-extremes

Your finalRatio equation is fine, but you must Synchonize() all the TimeSeries (or Indicators) first.

Where there gaps, Synchonize() will use the last known value.

QUOTE:
if I can limit max/min values in a timeseries.
Eg, I want all values in my timeseries that are > 100, to be changed to 100.

To do exactly that, you're going to have to loop through all the .Value's, but a better way to compression-limit the outliers would be to run them through the InverseFisher indicator. It uses a sigmoidal transfer function so there aren't any sharp cutoffs creating artifacts. Take a look at John Ehlers' paper. https://www.mesasoftware.com/papers/TheInverseFisherTransform.pdf

Of course, you should perform the synchronization discussed in Post #1 first, then linearly scale the values between +/- 1, and then call the InverseFisher indicator on each scaled time series to trim all the outliers. Finally, you can average them together.

On the +/- 1 scaling step, I would keep it simple.

scaledTimeSeries = unscaledTimeSeries / maxValueInTimeSeries;

Since I looked again and assuming all these variables are TimeSeries with the class scope, this should be done once in Initialize.

CODE:
   BINANCE_RATIO = TimeSeriesSynchronizer.Synchronize(BINANCE_RATIO, bars);
   BYBIT_RATIO_USDT = TimeSeriesSynchronizer.Synchronize(BYBIT_RATIO_USDT, bars);
   KRAKEN_RATIO = TimeSeriesSynchronizer.Synchronize(KRAKEN_RATIO, bars);
   BITMEX_RATIO = TimeSeriesSynchronizer.Synchronize(BITMEX_RATIO, bars);
   var finalRatio = (BINANCE_RATIO + BYBIT_RATIO_USDT + KRAKEN_RATIO + BITMEX_RATIO)/4;
   for (int n = 0; n < finalRatio.Count; n++)
   {
      if (finalRatio[n] > 100)
         finalRatio[n] = 100; 
   }