Introduction
There are multiple ways to smooth a series of data. I wrote an earlier entry detailing the use filters to smooth a data set. Kernel smoothing is another method that is related to filter smoothing. However, instead of using a given, limited set of weights to smooth the data, a function is used to calculate weights for every single data point.
I won't go into the mathematical detail here as Wikipedia has several pages that describe the method in a far better way than I ever could. All I will say is that the method is relatively simple. All we're doing is calculating a weighted average of each point based on every other single point. There are two pages that are helpful:
I won't go into the mathematical detail here as Wikipedia has several pages that describe the method in a far better way than I ever could. All I will say is that the method is relatively simple. All we're doing is calculating a weighted average of each point based on every other single point. There are two pages that are helpful:
- Kernel Smoother - describes the kernel smoother and how it works
- Kernel statistics - describes the various kernels that can be used
The Macro
The macro itself takes 4 inputs:
- The y-coordinates of the data series
- The x-coordinates of the data series
- The kernel to use
- The scale parameter of the kernel
The x and y coordinates are fairly self-explanatory. The x-coordinates are filled with values from 1 to the number of y values if no x-coordinates are specified.
The kernel is the function used to calculate the weight. There are 8 kernel functions to choose from:
The kernel is the function used to calculate the weight. There are 8 kernel functions to choose from:
- Gaussian
- Uniform
- Triangular
- Epanechnikov
- Quartic
- Cubic
- Logistic
- Cosine
The default kernel function used is a gaussian and the default scale is 1. The kernel function can be specified with either the full name or the first letter of the function (two in the case of the cosine function).
The scale parameter determines the 'width' of the function i.e. how many neighbouring points are used to calculate the new value of each point. The scale can either be a reference to a cell containing a number, a range or a number. If the scale is a range then it has to be the same size as the range of the y-coordinates if it's size is greater than 1 cell. Also the scales are used by applying scale 1 to point 1, scale 2 to point 2, etc. for every point.
The scale parameter determines the 'width' of the function i.e. how many neighbouring points are used to calculate the new value of each point. The scale can either be a reference to a cell containing a number, a range or a number. If the scale is a range then it has to be the same size as the range of the y-coordinates if it's size is greater than 1 cell. Also the scales are used by applying scale 1 to point 1, scale 2 to point 2, etc. for every point.
So the function takes the form:
=SmoothKernel(sy as Range,
Optional sx as Range,
Optional kerneltype as Variant,
Optional scaleP as Variant)
The Code
The code isn't particularly elegant, especially when it comes to specifying the scale parameter and the kernel function. I will probably try to change it at some point but it still works so maybe not.
Also the odd comments at the beginning of the macro are in the form they are because I created a macro to run through the functions and create some HTML documentation from them. I'll write an entry about it at some point.
Public Function SmoothKernel(sy As Range, Optional sx As Range, Optional kerneltype As Variant, Optional scaleP As Variant) As Variant
'@desc;Function to smooth data via kernels by converting every point to the corresponding kernel and summing the result.
'@param;sy;Range;The y coordinates
'@param;sx;Range;The x coordinates
'@param;kernel type;Variant;The kernel to apply. Should be one of Gaussian, Uniform, Triangular, Epanechnikov, Quartic, Cubic, Logistic or Cosine. Defaults to Gaussian.
'@param;scale;Variant;The size of the kernel
'@return;Variant;The smoothed points
Dim outputRows As Long
Dim outputCols As Long
Dim output() As Double
Dim vert As Boolean
Dim vertOutput As Boolean
Dim i As Integer
Dim j As Integer
Dim k As Integer
Dim x() As Double
Dim y() As Double
Dim kernel As String
Dim tot_ker1 As Double
Dim tot_ker2 As Double
Dim b() As Double
Dim temp_ker As Double
'Test whether data is arranged vertically or horizontally
vert = False
k = sy.Columns.Count
If sy.Rows.Count > sy.Columns.Count Then
vert = True
k = sy.Rows.Count
End If
With Application.Caller
outputRows = .Rows.Count
outputCols = .Columns.Count
End With
ReDim output(1 To outputRows, 1 To outputCols)
'Test whether output is arranged vertically or horizontally
vertOutput = False
If outputRows > outputCols Then
vertOutput = True
End If
'Check that output range is the same size as input range
If outputRows <> sy.Rows.Count And outputCols <> sy.Columns.Count Then Exit Function
'Populate output with zeroes
For i = 1 To outputRows
For j = 1 To outputCols
output(i, j) = 0
Next
Next
'Redimension variables
ReDim x(1 To k)
ReDim y(1 To k)
ReDim b(1 To k)
'Define kernels
If IsMissing(kerneltype) Then
kernel = "k_G"
ElseIf kerneltype = "Gaussian" Or kerneltype = "G" Then
kernel = "k_G"
ElseIf kerneltype = "Uniform" Or kerneltype = "U" Then
kernel = "k_U"
ElseIf kerneltype = "Triangular" Or kerneltype = "T" Then
kernel = "k_T"
ElseIf kerneltype = "Epanechnikov" Or kerneltype = "E" Then
kernel = "k_E"
ElseIf kerneltype = "Quartic" Or kerneltype = "Q" Then
kernel = "k_Q"
ElseIf kerneltype = "Cubic" Or kerneltype = "C" Then
kernel = "k_C"
ElseIf kerneltype = "Logistic" Or kerneltype = "L" Then
kernel = "k_L"
ElseIf kerneltype = "Cosine" Or kerneltype = "Co" Then
kernel = "k_Co"
Else
kernel = "k_G"
End If
'Define scale
If IsMissing(scaleP) Then
For i = 1 To k
b(i) = 1
Next
Else
If TypeName(scaleP) = "Range" Then
If scaleP.Rows.Count = 1 And scaleP.Columns.Count = 1 Then
For i = 1 To k
b(i) = scaleP
Next
ElseIf scaleP.Rows.Count <> sy.Rows.Count And scaleP.Columns.Count <> sy.Columns.Count Then
Exit Function
ElseIf scaleP.Rows.Count > scaleP.Columns.Count Then
For i = 1 To k
b(i) = scaleP(i, 1).Value
Next
Else
For i = 1 To k
b(i) = scaleP(1, i).Value
Next
End If
Else
For i = 1 To k
b(i) = scaleP
Next
End If
End If
'Populate temporary variables
For i = 1 To k
If sx Is Nothing Then
x(i) = i
Else
If vert = True Then
x(i) = sx(i, 1).Value
y(i) = sy(i, 1).Value
Else
x(i) = sx(1, i).Value
y(i) = sy(1, i).Value
End If
End If
Next
For i = 1 To k
tot_ker1 = 0
tot_ker2 = 0
For j = 1 To k
If kernel = "k_U" Then
temp_ker = k_U(x(i), x(j), b(j))
ElseIf kernel = "k_T" Then
temp_ker = k_T(x(i), x(j), b(j))
ElseIf kernel = "k_E" Then
temp_ker = k_E(x(i), x(j), b(j))
ElseIf kernel = "k_Q" Then
temp_ker = k_Q(x(i), x(j), b(j))
ElseIf kernel = "k_C" Then
temp_ker = k_C(x(i), x(j), b(j))
ElseIf kernel = "k_L" Then
temp_ker = k_L(x(i), x(j), b(j))
ElseIf kernel = "k_Co" Then
temp_ker = k_Co(x(i), x(j), b(j))
Else
temp_ker = k_G(x(i), x(j), b(j))
End If
tot_ker1 = tot_ker1 + temp_ker * y(j)
tot_ker2 = tot_ker2 + temp_ker
Next
If vertOutput = True Then
output(i, 1) = tot_ker1 / tot_ker2
Else
output(1, i) = tot_ker1 / tot_ker2
End If
Next
SmoothKernel = output
End Function
The kernel functions are:
Private Function k_G(x1 As Double, x2 As Double, b As Double) As Double
'Gaussian kernel
k_G = Exp(-(((x1 - x2) ^ 2) / (2 * b ^ 2)))
End Function
Private Function k_E(x1 As Double, x2 As Double, b As Double) As Double
'Epanechnikov kernel
If Abs((x1 - x2) / b) > 1 Then
k_E = 0
Else
k_E = (3 / 4) * (1 - ((x1 - x2) / b) ^ 2)
End If
End Function
Private Function k_L(x1 As Double, x2 As Double, b As Double) As Double
'Logistic kernel
k_L = 1 / (Exp((x1 - x2) / b) + Exp(-(x1 - x2) / b))
End Function
Private Function k_U(x1 As Double, x2 As Double, b As Double) As Double
'Uniform kernel
If Abs((x1 - x2) / b) > 1 Then
k_U = 0
Else
k_U = 1 / 2
End If
End Function
Private Function k_T(x1 As Double, x2 As Double, b As Double) As Double
'Triangular kernel
If Abs((x1 - x2) / b) > 1 Then
k_T = 0
Else
k_T = (1 - Abs((x1 - x2) / b))
End If
End Function
Private Function k_Q(x1 As Double, x2 As Double, b As Double) As Double
'Quartic kernel
If Abs((x1 - x2) / b) > 1 Then
k_Q = 0
Else
k_Q = (15 / 16) * (1 - ((x1 - x2) / b) ^ 2) ^ 2
End If
End Function
Private Function k_C(x1 As Double, x2 As Double, b As Double) As Double
'Cubic kernel
If Abs((x1 - x2) / b) > 1 Then
k_C = 0
Else
k_C = (35 / 32) * (1 - ((x1 - x2) / b) ^ 2) ^ 3
End If
End Function
Private Function k_Co(x1 As Double, x2 As Double, b As Double) As Double
'Cosine kernel
If Abs((x1 - x2) / b) > 1 Then
k_Co = 0
Else
k_Co = (WorksheetFunction.Pi / 4) * Cos((WorksheetFunction.Pi / 2) * ((x1 - x2) / b))
End If
End Function
Example
Just to give you a flavour of what the macro does, I created a series of data based on the sine function. Specifically the data is sin(x)+0.3*rand()-0.3
This will also show the shortcomings of the method and one way of surmounting them (although not entirely).
The above picture shows the original data series with the s=axis showing degrees. Let's apply the function to the data by entering:
This will also show the shortcomings of the method and one way of surmounting them (although not entirely).
=SmoothKernel(C2:C362,A2:A362,"G",1)
into cells D2:D362 and then pressing Ctrl+Shift+Enter to enter the formula. This gives:
It's better but not smooth. I've also added a dashed line to indicate a sine function - the original function that we hope to obtain from smoothing the data. So now can increase the scale to 7:
This makes the resulting data series smooth but notice the ends. Because there are no data points to the left of the first few points (obviously) the estimate is too high. It's correct in the sense that this is the weighted average for the point but it's obviously not a good estimate.
So what can we do? Change the scale parameter for the points around the ends. If we increase the scale where the estimate is off and decrease the scale close to these points. In this case, I made the weights 14, 12, 10 and then 3 for 19 cells. This gave:
Better, but it still shows end effects so you need to be careful what scale/weight you apply.
One other way of applying the weights here would be to apply weight 1 to all points when calculating the smoothed value of point 1. This would limit the end effect in a different way. I haven't written this into the macro yet but may do later.
As ever comments are welcome.