Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

Let X_1, X_2, ..., X_n, and Y_1, Y_2, ..., Y_m be independent random variables,

ID: 3227135 • Letter: L

Question

Let X_1, X_2, ..., X_n, and Y_1, Y_2, ..., Y_m be independent random variables, with the X' s being a random sample from a N (mu_x, sigma^2_x) distribution, and the Y' s being a random sample from a N (mu_y, sigma^2_Y) distribution. Suppose that all four parameters, mu_x, sigma^2_X, mu_Y, sigma^2_Y, are unknown, but also suppose that it is known that sigma^2_X = 5 sigma^2_Y. Use this information to develop a t - ratio that has n + m - 2 degrees of freedom, and use that to find a formula for a 100 (1 - alpha) % confidence interval for mu_X - mu_Y.

Explanation / Answer

Point estimate for 100(1-alpha)% CI=

(MuX-MuY)

pooled std deviation Sp= sqrt(((n-1)Sx^2-(m-1)Sy^2)/(n+m-2))

{now using given information Sx^2 = 5Sy^2

Sp= sqrt(((n-1)Sx^2-5(m-1)Sx^2)/(n+m-2)) = Sx* sqrt((n-5m+4)/(n+m-2))

pooled std error= Sp*sqrt(1/n+1/m) = Sx* sqrt(((n-5m+4)/(n+m-2))*(1/n+1/m))

thus,

100(1-alpha)% CI:

(MuX-MuY) +- t * Std error=

=(MuX-MuY) +- t*Sx* sqrt(((n-5m+4)/(n+m-2))*(1/n+1/m))