Normalized MACD

[KlekotOliwa]

in Skender.Stock.Indicators I miss normalization
a_madcList = quotesPredicationList.GetMacd().RemoveWarmupPeriods().ToList();
a_madc = a_madcList.GetMacd().Last();
a_madcNormalized = new MacdResult(a_madc.Date);// { Macd = a_madc.Macd, Signal = a_madc.Signal, Histogram = a_madc.Histogram, FastEma= a_madc.FastEma, SlowEma= a_madc.SlowEma}; //Helper.CreateDeepCopy(a_madc);
a_madcNormalized.Macd = (float)_util.Normalize((decimal)a_madc.Macd, (decimal)a_madcList.Min(m => m.Macd), (decimal)a_madcList.Max(m => m.Macd), -1000, 1000);
a_madcNormalized.Signal = (float)_util.Normalize((decimal)a_madc.Signal, (decimal)a_madcList.Min(m => m.Signal), (decimal)a_madcList.Max(m => m.Signal), -1000, 1000);
a_madcNormalized.Histogram = (float)_util.Normalize((decimal)a_madc.Histogram, (decimal)a_madcList.Min(m => m.Histogram), (decimal)a_madcList.Max(m => m.Histogram), -1000, 1000);
a_madcNormalized.FastEma = (float)_util.Normalize((decimal)a_madc.FastEma, (decimal)a_madcList.Min(m => m.FastEma), (decimal)a_madcList.Max(m => m.FastEma), -1000, 1000);
a_madcNormalized.SlowEma = (float)_util.Normalize((decimal)a_madc.SlowEma, (decimal)a_madcList.Min(m => m.SlowEma), (decimal)a_madcList.Max(m => m.SlowEma), -1000, 1000);

[DaveSkender]

in Skender.Stock.Indicators I miss normalization

I'm doing a quick reformat-only of your code so I can visualize it better.

// calculate regular MACD
var a_madcList = quotesPredicationList
  .GetMacd()
  .RemoveWarmupPeriods()
  .ToList();

// calculate a MACD of the MACD and
// fetch its last 'MacdResult' entry
var a_madc = a_madcList.GetMacd().Last();

// create a copy of it for normalized conversion,
// to avoid using ref of 'a_macd'
var a_madcNormalized = new MacdResult(a_madc.Date) { 
  Macd      = a_madc.Macd,
  Signal    = a_madc.Signal,
  Histogram = a_madc.Histogram,
  FastEma   = a_madc.FastEma,
  SlowEma   = a_madc.SlowEma
};

// recalculate 'normalized' property values

// for MACD
a_madcNormalized.Macd = (float)_util.Normalize(
  (decimal)a_madc.Macd,
  (decimal)a_madcList.Min(m => m.Macd), 
  (decimal)a_madcList.Max(m => m.Macd),
  -1000, 1000);

// for Signal
a_madcNormalized.Signal = (float)_util.Normalize(
  (decimal)a_madc.Signal,
  (decimal)a_madcList.Min(m => m.Signal),
  (decimal)a_madcList.Max(m => m.Signal),
  -1000, 1000);

// for Histogram
a_madcNormalized.Histogram = (float)_util.Normalize(
  (decimal)a_madc.Histogram,
  (decimal)a_madcList.Min(m => m.Histogram),
  (decimal)a_madcList.Max(m => m.Histogram),
  -1000, 1000);

// for FastEma
a_madcNormalized.FastEma = (float)_util.Normalize(
  (decimal)a_madc.FastEma,
  (decimal)a_madcList.Min(m => m.FastEma),
  (decimal)a_madcList.Max(m => m.FastEma),
  -1000, 1000);

// for SlowEma
a_madcNormalized.SlowEma = (float)_util.Normalize(
  (decimal)a_madc.SlowEma,
  (decimal)a_madcList.Min(m => m.SlowEma),
  (decimal)a_madcList.Max(m => m.SlowEma),
  -1000, 1000);

From what I'm seeing, I suspect you're trying to create a bounded oscillator version of MACD by using a custom normalization method with a signature like this:

public decimal Normalize(
  decimal value,
  decimal minValue,
  decimal maxValue,
  int lowerBoundary,
  int upperBoundary) {

  // calculate normalized value based on lower/upper boundaries

}

I'm not sure if you're proposing that we add this Normalize() utility to the library, seeking help/support in understanding what you're doing, or if you're just sharing an interesting approach to the community.

What I can tell you, if you're looking for an interesting related fact about MACD-based bounded oscillators, is that the Schaff Trend Cycle (STC) is just a Stochastic Oscillator of a Moving Average Convergence / Divergence (MACD) indicator.

[KlekotOliwa]

I have included the topic of normalization in this section because I am missing this basic method of organizing data in Skender.Stock.Indicators

[DaveSkender]

There are many types and uses for normalization of data. I think what you are doing here is a form of feature scaling.

If you are using different base scale for each of the MACD components, in this case, you’re probably changing the relative meaning it. For example, histogram may no longer be the difference between MACD and Signal line.