Project 47: Standard Error of Slope
yi = the y-value for the point with index i
xi = the x-value for the point with index i
ŷi = the predicted y-value for the point with index i
x̄ = the mean value for the x-values
Example:
x = [1, 3, 5]
y = [5, 7, 12]
xbar = 3
LSRL: y = 1.75x + 2.75
predicted ys = 1.75(1) + 2.75, 1.75(3) + 2.75, 1.75(5) + 2.75 = 4.5, 8, 11.5
residuals = .5, -1, .5
sum of squared residuals = .25 + 1 + .25 = 1.5
differences from xbar = -2, 0, 2
sum of squared differences from xbar = 4 + 0 + 4 = 8
seSlope = sqrt(1/1*1.5/8) = .433
Project 47: Variables 'x' and 'y' have been initialized. There are also working methods called getAverage and residual.
x and y represent the x and y values in the dataset.
getAverage(data) returns the average of data
residual(x, y, index) returns the residual for the point with index given
Example: If x = [1, 3, 5] and y = [5, 7, 12] then residual(x, y, 2) returns .5 because the residual for the point (5,12) is .5.
Task: Appropriately initialize the value of 'seSlope' that represents the standard error of the slope (or the approximate standard deviation of the sampling distribution of sample slopes)
**If your code works for 5 test cases, you can enter your e-mail address
Universal Computational Math Methods:
pow(5,2) returns 25.0
abs(-3.0) returns 3
sqrt(49.0) returns 7.0