Showing posts with label Correlation and Regression. Show all posts
Showing posts with label Correlation and Regression. Show all posts

Correlation and Regression

The data for marks in Physics and History obtained by $10$ students is as given:

Marks in Physics $15$ $12$ $8$ $8$ $7$ $7$ $7$ $6$ $5$ $3$
Marks in History $10$ $25$ $17$ $11$ $13$ $17$ $20$ $13$ $9$ $15$

Using this data, compute the Karl Pearson's coefficient of correlation between the marks in Physics and History obtained by the $10$ students.


Number of students $= n = 10$

Marks in Physics $\left(x\right)$ Marks in History $\left(y\right)$
$15$ $10$
$12$ $25$
$8$ $17$
$8$ $11$
$7$ $13$
$7$ $17$
$7$ $20$
$6$ $13$
$5$ $9$
$3$ $15$
$\Sigma x = 78$ $\Sigma y = 150$


$\overline{x} = \dfrac{\Sigma x}{n} = \dfrac{78}{10} = 7.8$, $\;$ $\overline{y} = \dfrac{\Sigma y}{n} = \dfrac{150}{10} = 15$

Marks in Physics $\left(x\right)$ Marks in History $\left(y\right)$ $d_x = x - \overline{x}$ $d_y = y - \overline{y}$ $d_x \cdot d_y$ $d_x^2$ $d_y^2$
$15$ $10$ $7.2$ $- 5$ $- 36$ $51.84$ $25$
$12$ $25$ $4.2$ $10$ $42$ $17.64$ $100$
$8$ $17$ $0.2$ $2$ $0.4$ $0.04$ $4$
$8$ $11$ $0.2$ $- 4$ $- 0.8$ $0.04$ $16$
$7$ $13$ $- 0.8$ $- 2$ $1.6$ $0.64$ $4$
$7$ $17$ $- 0.8$ $2$ $- 1.6$ $0.64$ $4$
$7$ $20$ $- 0.8$ $5$ $- 4$ $0.64$ $25$
$6$ $13$ $- 1.8$ $- 2$ $3.6$ $3.24$ $4$
$5$ $9$ $- 2.8$ $- 6$ $16.8$ $7.84$ $36$
$3$ $15$ $- 4.8$ $0$ $0$ $23.04$ $0$
$\Sigma x = 78$ $\Sigma y = 150$ $\Sigma d_x d_y = 22$ $\Sigma d_x^2 = 82.56$ $\Sigma d_y^2 = 218$


Karl Pearson coefficient $= r = \dfrac{\Sigma d_x d_y}{\sqrt{\Sigma d_x^2 \cdot \Sigma d_y^2}}$

i.e. $\;$ $r = \dfrac{22}{\sqrt{82.56 \times 218}} = \dfrac{22}{\sqrt{17998.08}} = \dfrac{22}{134.16} = 0.1640$ (approx)

Correlation and Regression

Find the regression coefficients $b_{yx}$ and $b_{xy}$ for the following data:
$n = 6$, $\;$ $\Sigma x = 30$, $\;$ $\Sigma y = 42$, $\;$ $\Sigma xy = 199$, $\;$ $\Sigma x^2 = 184$, $\;$ $\Sigma y^2 = 318$


$\begin{aligned} b_{yx} & = \dfrac{\Sigma xy - \dfrac{\Sigma x \Sigma y}{n}}{\Sigma x^2 - \dfrac{\left(\Sigma x\right)^2}{n}} \\\\ & = \dfrac{199 - \dfrac{30 \times 42}{6}}{184 - \dfrac{\left(30\right)^2}{6}} \\\\ & = \dfrac{199 - 210}{184 - 150} = \dfrac{-11}{34} = - 0.3235 \end{aligned}$

$\begin{aligned} b_{xy} & = \dfrac{\Sigma xy - \dfrac{\Sigma x \Sigma y}{n}}{\Sigma y^2 - \dfrac{\left(\Sigma y\right)^2}{n}} \\\\ & = \dfrac{199 - \dfrac{30 \times 42}{6}}{318 - \dfrac{\left(42\right)^2}{6}} \\\\ & = \dfrac{199 - 210}{318 - 294} = \dfrac{-11}{24} = - 0.4583 \end{aligned}$