Class: Statsample::Crosstab

Inherits:
Object show all
Includes:
Summarizable
Defined in:
lib/statsample/crosstab.rb

Overview

Class to create crosstab of data With this, you can create reports and do chi square test The first vector will be at rows and the second will the the columns

Instance Attribute Summary collapse

Instance Method Summary collapse

Methods included from Summarizable

#summary

Constructor Details

#initialize(v1, v2, opts = Hash.new) ⇒ Crosstab

Returns a new instance of Crosstab.

Raises:

  • (ArgumentError) 


10
11
12
13
14
15
16
17
18
19
20
21
22
23
 
# File 'lib/statsample/crosstab.rb', line 10

def initialize(v1, v2, opts=Hash.new)
  #raise ArgumentError, "Both arguments should be Vectors" unless v1.is_a? Statsample::Vector and v2.is_a? Statsample::Vector
  raise ArgumentError, "Vectors should be the same size" unless v1.size==v2.size
  @v_rows, @v_cols=Statsample.only_valid_clone(v1.to_vector,v2.to_vector)
  @cases=@v_rows.size
  @row_label=v1.name
  @column_label=v2.name
  @name=nil
  @percentage_row = @percentage_column = @percentage_total=false
  opts.each{|k,v|
    self.send("#{k}=",v) if self.respond_to? k
  }
  @name||=_("Crosstab %s - %s") % [@row_label, @column_label]
end

Instance Attribute Details

#column_label ⇒ Object

Returns the value of attribute column_label.



9
10
11
 
# File 'lib/statsample/crosstab.rb', line 9

def column_label
  @column_label
end

#name ⇒ Object

Returns the value of attribute name.



9
10
11
 
# File 'lib/statsample/crosstab.rb', line 9

def name
  @name
end

#percentage_column ⇒ Object

Returns the value of attribute percentage_column.



9
10
11
 
# File 'lib/statsample/crosstab.rb', line 9

def percentage_column
  @percentage_column
end

#percentage_row ⇒ Object

Returns the value of attribute percentage_row.



9
10
11
 
# File 'lib/statsample/crosstab.rb', line 9

def percentage_row
  @percentage_row
end

#percentage_total ⇒ Object

Returns the value of attribute percentage_total.



9
10
11
 
# File 'lib/statsample/crosstab.rb', line 9

def percentage_total
  @percentage_total
end

#row_label ⇒ Object

Returns the value of attribute row_label.



9
10
11
 
# File 'lib/statsample/crosstab.rb', line 9

def row_label
  @row_label
end

#v_cols ⇒ Object (readonly)

Returns the value of attribute v_cols.



8
9
10
 
# File 'lib/statsample/crosstab.rb', line 8

def v_cols
  @v_cols
end

#v_rows ⇒ Object (readonly)

Returns the value of attribute v_rows.



8
9
10
 
# File 'lib/statsample/crosstab.rb', line 8

def v_rows
  @v_rows
end

Instance Method Details

#chi_square ⇒ Object

Chi square, based on expected and real matrix



69
70
71
72
 
# File 'lib/statsample/crosstab.rb', line 69

def chi_square
    require 'statsample/test'
    Statsample::Test.chi_square(self.to_matrix, matrix_expected)
end

#cols_empty_hash ⇒ Object 



87
88
89
 
# File 'lib/statsample/crosstab.rb', line 87

def cols_empty_hash
  cols_names.inject({}) {|a,x| a[x]=0;a}
end

#cols_names ⇒ Object 



27
28
29
 
# File 'lib/statsample/crosstab.rb', line 27

def cols_names
  @v_cols.factors.sort
end

#cols_total ⇒ Object 



33
34
35
 
# File 'lib/statsample/crosstab.rb', line 33

def cols_total
  @v_cols.frequencies
end

#frequencies ⇒ Object 



37
38
39
40
41
42
43
44
45
 
# File 'lib/statsample/crosstab.rb', line 37

def frequencies
  base=rows_names.inject([]){|s,row| 
    s+=cols_names.collect{|col| [row,col]}
  }.inject({}) {|s,par|
    s[par]=0
    s
  }
  base.update(Statsample::vector_cols_matrix(@v_rows,@v_cols).to_a.to_vector.frequencies)
end

#frequencies_by_col ⇒ Object 



61
62
63
64
65
66
67
 
# File 'lib/statsample/crosstab.rb', line 61

def frequencies_by_col
  f=frequencies
  cols_names.inject({}){|sc,col| 
    sc[col]=rows_names.inject({}) {|sr,row| sr[row]=f[[row,col]]; sr}
    sc
  }
end

#frequencies_by_row ⇒ Object 



54
55
56
57
58
59
60
 
# File 'lib/statsample/crosstab.rb', line 54

def frequencies_by_row
f=frequencies
rows_names.inject({}){|sr,row|
  sr[row]=cols_names.inject({}) {|sc,col| sc[col]=f[[row,col]]; sc}
  sr
}
end

#matrix_expected ⇒ Object

Useful to obtain chi square



74
75
76
77
78
79
80
81
82
83
84
85
86
 
# File 'lib/statsample/crosstab.rb', line 74

def matrix_expected
  rn=rows_names
  cn=cols_names
  rt=rows_total
  ct=cols_total
  t=@v_rows.size
  m=rn.collect{|row|
    cn.collect{|col|
      (rt[row]*ct[col]).quo(t) 
      }
  }
  Matrix.rows(m)
end

#report_building(builder) ⇒ Object 



90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
 
# File 'lib/statsample/crosstab.rb', line 90

def report_building(builder)
  builder.section(:name=>@name) do |generator|
    fq=frequencies
    rn=rows_names
    cn=cols_names
    total=0
    total_cols=cols_empty_hash
    generator.text "Chi Square: #{chi_square}"
    generator.text(_("Rows: %s") % @row_label) unless @row_label.nil?
    generator.text(_("Columns: %s") % @column_label) unless @column_label.nil?
    
    t=ReportBuilder::Table.new(:name=>@name+" - "+_("Raw"), :header=>[""]+cols_names.collect {|c| @v_cols.labeling(c)}+[_("Total")])
    rn.each do |row|
      total_row=0
      t_row=[@v_rows.labeling(row)]
      cn.each do |col|
        data=fq[[row,col]]
        total_row+=fq[[row,col]]
        total+=fq[[row,col]]                    
        total_cols[col]+=fq[[row,col]]                    
        t_row.push(data)
      end
      t_row.push(total_row)
      t.row(t_row)
    end
    t.hr
    t_row=[_("Total")]
    cn.each do |v|
      t_row.push(total_cols[v])
    end
    t_row.push(total)
    t.row(t_row)
    generator.parse_element(t)
    
    if(@percentage_row)
      table_percentage(generator,:row)
    end
    if(@percentage_column)
    table_percentage(generator,:column)
    end
    if(@percentage_total)
    table_percentage(generator,:total)
    end
  end
end

#rows_names ⇒ Object 



24
25
26
 
# File 'lib/statsample/crosstab.rb', line 24

def rows_names
  @v_rows.factors.sort
end

#rows_total ⇒ Object 



30
31
32
 
# File 'lib/statsample/crosstab.rb', line 30

def rows_total
  @v_rows.frequencies
end

#table_percentage(generator, type) ⇒ Object 



138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
 
# File 'lib/statsample/crosstab.rb', line 138

def table_percentage(generator,type)
  fq=frequencies
  cn=cols_names
  rn=rows_names
  rt=rows_total
  ct=cols_total
  
  type_name=case type
    when :row     then  _("% Row")
    when :column  then  _("% Column")
    when :total   then  _("% Total")
  end
  
  t=ReportBuilder::Table.new(:name=>@name+" - "+_(type_name), :header=>[""]+cols_names.collect {|c| @v_cols.labeling(c) } + [_("Total")])
    rn.each do |row|
      t_row=[@v_rows.labeling(row)]
      cn.each do |col|
        total=case type
          when :row     then  rt[row]
          when :column  then  ct[col]
          when :total   then  @cases
        end
        data = sprintf("%0.2f%%", fq[[row,col]]*100.0/ total )
        t_row.push(data)
      end
      total=case type
        when :row     then  rt[row]
        when :column  then  @cases
        when :total   then  @cases
      end              
      t_row.push(sprintf("%0.2f%%", rt[row]*100.0/total))
      t.row(t_row)
    end
    
    t.hr
    t_row=[_("Total")]
    cn.each{|col|
      total=case type
        when :row     then  @cases
        when :column  then  ct[col]
        when :total   then  @cases
      end
      t_row.push(sprintf("%0.2f%%", ct[col]*100.0/total))
    }
  t_row.push("100%")
  t.row(t_row)
  generator.parse_element(t)
end

#to_matrix ⇒ Object 



46
47
48
49
50
51
52
53
 
# File 'lib/statsample/crosstab.rb', line 46

def to_matrix
  f=frequencies
  rn=rows_names
  cn=cols_names
  Matrix.rows(rn.collect{|row|
      cn.collect{|col| f[[row,col]]}
  })
end