Formulas and Output
\[Coefficient = \frac{\sum XY - (\sum X)(\sum Y)/n}{\sum X^2-\sum (X)^2/n}=\frac{\ 22126 - 434\times445/10}{ 22426 - 188356/10} = 0.7835\]
\[Intercept = \bar{y}- b \times\bar{x} = 44.5- (0.7835 \times 43.4) = 10.497\]
\[SSR = b\sum XY - (\sum X)(\sum Y)/n = 0.7835 \times 22126 \times 445 / 10 = 2203.9241
\]
\[SSE = SST - SSR = 2300.5 - 2203.9241 = 96.5759\]
\[SST = \sum Y^2 - (\sum Y)^2/n = 22103 - 445^2 / 10 = 2300.5
\]
\[MSR = SSR / 1 = 2203.9241\]
\[MSE = SSE / n-2 = 96.5759/(10-2) = 12.072\]
Example
Robert E. Moritz wants to find out using a simple linear regression model if the accountants
age (X)
has an effect on the time (in minutes) to finish an accounting task (Y).
The following data is from a sample of 10 accountants:
X = [20 25 30 37 45 47 55 59 62 65]
Y = [15 17 25 32 51 43 60 65 58 68]
a) Find the intercept and the coefficient.
b) Set up the ANOVA table and test if the coefficient is significant at p=0.05.
Using this calculator you can find the answer to the questions above
on any simple regression dataset and the method used to find them.