Page 292 - Artificial Intellegence_v2.0_Class_11
P. 292

N = No. of samples = 5
                        m =   N Σ(xy) − Σx Σy
                              N Σ(x ) − (Σx) 2
                                  2
                             5(64) – (10) (25)
                            =
                              5 (30) – (10) 2

                             320 – 250
                            =
                             150 – 100
                            70
                            =
                            50

                             Σy − m Σx
                            b =
                                 N
                          25 – 1.4 (10)
                         =            = 2.2
                               5

                        Putting the values in the equation y = mx + b, we get the regression equation
                        y = 1.4 x + 2.2
                        b.   When x = 10,

                             y = 1.4 (10) + 2.2  =  16.2
                     2.  Describe the four assumptions for Pearson's correlation coefficient.
                   Ans.  There  are  some  assumptions  for  Pearson's  correlation  coefficient  which  are  as  follows.  If  any  of  these  four
                        requirements are not met, analysis of data using Pearson's correlation coefficient might not yield a valid result.
                        (a)     The data type of the two variables should be continuous.
                        (b)     There  must  be  a  linear  relationship  between  the  two  variables.  Create  a  scatterplot  by  plotting  the  two
                             variables against each other. The scatterplot can then be used to check for linearity.
                        (c)     The data should not have any significant outliers. Outliers can have a great impact on the line of best fit and
                             the Pearson correlation coefficient, leading to very difficult inferences regarding the data. Hence, it is best to
                             have no outliers or keep them to a minimum.
                        (d)   The variables should be normally distributed (approximately).

                     3.  The table below shows the crosstab of gender and highest education degree achieved by 500 people in a survey:
                             No Secondary     Secondary     Higher Secondary    Graduation   Post Graduation  Total
                             School Degree   School Degree    School Degree      Degree         Degree
                    Males          50             95               11               39            23         218
                    Females        52            166               15               37            12         282
                    Total         102            261               26              76             35         500

              Calculate:
                   a.   What percent of the sample were males?
                   Ans.  218/500 = 43.6%

                   b.   What percent of the sample were males with no secondary school degree?
                   Ans.  50/500 = 10%

                   c.   What percent of the people did not graduate from secondary school?
                   Ans.  102/500 = 20.4%



                    290     Touchpad Artificial Intelligence (Ver. 2.0)-XI
   287   288   289   290   291   292   293   294   295   296   297