Top 20% 5 minute up and down moves over the last 10 days

Author: cfurer

Creation Date: 6/9/2010 9:43 PM

Wealth Lab » WealthScript (.NET) » Top 20% 5 minute up and down moves over the last 10 days

cfurer

6/9/2010 9:43 PM

Hi. I want to do the following:
* Step through the 5 minute bars over the last 10 days.
* For each 5 minute bar calculate the difference between each bar's open & close, which I'll call the up-move (C>O) and down-move (C<O).
* Segregate the top 20% up-moving bars and down-moving bars.
* Calculate separate averages of the top 20% up-moves and down-moves.

I will plot these averages on daily and intraday charts to identify divergences between the longer price trend and the average size of the up & down 5min moves.

I'm stumped on how to sort the values in order to segragate the top 20% up & down moves. Is there any code out there that performs this kind of logic that I can use as a template?

Thanks

Eugene

6/10/2010 4:06 AM

1. Put the up-moving and down-moving values into two lists i.e. List<double>.
2. Determine the 80th percentile (i.e. 1 - 20/100) for each list (see function below). These are the top 20% values.
3. Then use a delegate to search your Lists for values greater or equal than the percentile, and you'll have two new List<double> objects containing only the Top-20 values. Here's an example using one list:

CODE:
Please log in to see this code.

Finally, here's the percentile function:

CODE:
Please log in to see this code.

Size:

Color:

1. Put the up-moving and down-moving values into two lists i.e. [i]List<double>[/i].
2. Determine the 80th percentile (i.e. 1 - 20/100) for each list (see function below). These are the top 20% values.
3. Then use a delegate to search your Lists for values greater or equal than the percentile, and you'll have two new [i]List<double>[/i] objects containing only the Top-20 values. Here's an example using one list:
[CODE]List<double> In = new List<double>();
In.Add(10); In.Add(100); In.Add(15);
			
double Top20 = 25;
			
List<double> Out = In.FindAll(delegate(double d) { return d >= Top20; });
foreach( double d in Out )
	PrintDebug( d );[/CODE]

Finally, here's the percentile function:
[CODE]// http://www.codeproject.com/KB/recipes/DescriptiveStatisticClass.aspx
// modified by Eugene to work with Lists
/// <summary>
/// Calculate percentile of a sorted data set
/// </summary>
/// <param name="sortedData"></param>
/// <param name="p"></param>
/// <returns></returns>
public double percentile(List<double> sortedData, double p)
{
	// algo derived from Aczel pg 15 bottom
	if (p >= 100.0d) return sortedData[sortedData.Count - 1];

double position = (double)(sortedData.Count + 1) * p / 100.0;
	double leftNumber = 0.0d, rightNumber = 0.0d;

double n = p / 100.0d * (sortedData.Count - 1) + 1.0d;

if (position >= 1)
	{
		leftNumber = sortedData[(int)System.Math.Floor(n) - 1];
		rightNumber = sortedData[(int)System.Math.Floor(n)];
	}
	else
	{
		leftNumber = sortedData[0]; // first data
		rightNumber = sortedData[1]; // first data
	}

if (leftNumber == rightNumber)
		return leftNumber;
	else
	{
		double part = n - System.Math.Floor(n);
		return leftNumber + part * (rightNumber - leftNumber);
	}
}[/CODE]