public override void HandleNextTick(Tick t)
{
double emaVal = EMA.HandleNextTick(t);
LRValues.Enqueue(emaVal);
double[] Y = LRValues.ToArray();
LinearRegression l = new LinearRegression();
l.Model(Y);
HighData.Enqueue(EMA[0]+STDEV * l.STDERR);
MiddleData.Enqueue(EMA[0]);
LowData.Enqueue(EMA[0] - STDEV * l.STDERR);
}
public static double CalculateHurstEstimate(double[] data, int minLength = 8)
{
List<double> n = new List<double>();
List<double> avgRange = new List<double>();
for (int i = 1; i <= data.Length; i *= 2)
{
double range = CalculateAverageRange(data, data.Length/i);
if (!range.Equals(double.NaN))
{
n.Add(i);
avgRange.Add(range);
}
}
LinearRegression LR = new LinearRegression();
LR.Model(n.Select(i=> Math.Log(i)).ToArray(),avgRange.Select(i=> Math.Log(i)).ToArray());
return LR.X2;
}
public static double CalculateHurstEstimate(double[] data, int minLength = 8)
{
List <double> n = new List <double>();
List <double> avgRange = new List <double>();
for (int i = 1; i <= data.Length; i *= 2)
{
double range = CalculateAverageRange(data, data.Length / i);
if (!range.Equals(double.NaN))
{
n.Add(i);
avgRange.Add(range);
}
}
LinearRegression LR = new LinearRegression();
LR.Model(n.Select(i => Math.Log(i)).ToArray(), avgRange.Select(i => Math.Log(i)).ToArray());
return(LR.X2);
}
public override double HandleNextTick(Tick t)
{
double value = double.NaN;
MA.HandleNextTick(t);
indicator.HandleNextTick(t);
if (!MA[0].Equals(double.NaN))
{
double[] MAArray = MA.ToArray();
double[] indArray = indicator.ToArray();
SortedList<double, double> tickHighs = new SortedList<double, double>();
SortedList<double, double> tickLows = new SortedList<double, double>();
SortedList<double, double> indHighs = new SortedList<double, double>();
SortedList<double, double> indLows = new SortedList<double, double>();
for (int i = 2; i < MAArray.Length; i++)
{
//high
if (MAArray[i] < MAArray[i - 1] && MAArray[i - 1] > MAArray[i - 2])
{
tickHighs.Add(i - 1, MAArray[i - 1]);
}
if (indArray[i] < indArray[i - 1] && indArray[i - 1] > indArray[i - 2])
{
indHighs.Add(i - 1, indArray[i - 1]);
}
//low
if (MAArray[i] > MAArray[i - 1] && MAArray[i - 1] < MAArray[i - 2])
{
tickLows.Add(i - 1, MAArray[i - 1]);
}
if (indArray[i] > indArray[i - 1] && indArray[i - 1] < indArray[i - 2])
{
indLows.Add(i - 1, indArray[i - 1]);
}
}
if (tickHighs.Count > 0 && indHighs.Count > 0
&& tickLows.Count > 0 && indLows.Count > 0)
{
LinearRegression lrTickHighs = new LinearRegression();
LinearRegression lrTickLows = new LinearRegression();
LinearRegression lrindkHighs = new LinearRegression();
LinearRegression lrindLows = new LinearRegression();
lrTickHighs.Model(tickHighs.Keys.ToArray(), tickHighs.Values.ToArray());
lrTickLows.Model(tickLows.Keys.ToArray(), tickLows.Values.ToArray());
lrindkHighs.Model(indHighs.Keys.ToArray(), indHighs.Values.ToArray());
lrindLows.Model(indLows.Keys.ToArray(), indLows.Values.ToArray());
if (lrTickHighs.X2 > 0 && lrindkHighs.X2 < 0) value = 1;
else if (lrTickLows.X2 < 0 && lrindLows.X2 > 0) value = -1;
else value = 0;
value = lrindkHighs.Fit(indArray[0]);
}
}
indicatorData.Enqueue(value);
return value;
}