Class: MoreMath::LinearRegression

Inherits:
Object
  • Object
show all
Defined in:
lib/more_math/linear_regression.rb

Overview

This class computes a linear regression for the given image and domain data sets.

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(image, domain = (0...image.size).to_a) ⇒ LinearRegression

Returns a new instance of LinearRegression.



5
6
7
8
9
10
 
# File 'lib/more_math/linear_regression.rb', line 5

def initialize(image, domain = (0...image.size).to_a)
  image.size != domain.size and raise ArgumentError,
    "image and domain have unequal sizes"
  @image, @domain = image, domain
  compute
end

Instance Attribute Details

#a ⇒ Object (readonly)

The slope of the line.



19
20
21
 
# File 'lib/more_math/linear_regression.rb', line 19

def a
  @a
end

#b ⇒ Object (readonly)

The offset of the line.



22
23
24
 
# File 'lib/more_math/linear_regression.rb', line 22

def b
  @b
end

#domain ⇒ Object (readonly)

The domain data as an array.



16
17
18
 
# File 'lib/more_math/linear_regression.rb', line 16

def domain
  @domain
end

#image ⇒ Object (readonly)

The image data as an array.



13
14
15
 
# File 'lib/more_math/linear_regression.rb', line 13

def image
  @image
end

Instance Method Details

#r2 ⇒ Object 



45
46
47
48
49
50
51
52
 
# File 'lib/more_math/linear_regression.rb', line 45

def r2
  image_seq = MoreMath::Sequence.new(@image)
  sum_res   = residuals.inject(0.0) { |s, r| s + r ** 2 }
  [
    1.0 -  sum_res / image_seq.sum_of_squares,
    0.0,
  ].max
end

#residuals ⇒ Object

Returns the residuals of this linear regression in relation to the given domain and image.



37
38
39
40
41
42
43
 
# File 'lib/more_math/linear_regression.rb', line 37

def residuals
  result = []
  @domain.zip(@image) do |x, y|
    result << y - (@a * x + @b)
  end
  result
end

#slope_zero?(alpha = 0.05) ⇒ Boolean

Return true if the slope of the underlying data (not the sample data passed into the constructor of this LinearRegression instance) is likely (with alpha level alpha) to be zero.

Returns:

  • (Boolean) 


27
28
29
30
31
32
33
 
# File 'lib/more_math/linear_regression.rb', line 27

def slope_zero?(alpha = 0.05)
  df = @image.size - 2
  return true if df <= 0 # not enough values to check
  t = tvalue(alpha)
  td = TDistribution.new df
  t.abs <= td.inverse_probability(1 - alpha.abs / 2.0).abs
end