I have conflicting web-based references for the following problem: If a response variable y is normally distributed, and so: Beta(be) ~ N(Beta,sigma^2/Sxx)
Beta(be)-Beta / [sigma/sq(Sxx)] ~ N(0,1) Where: Beta = slope of least square fitted line (be) = best estimate Sxx = sum of squares of x With: sigma^2=var(y) will be unknown can estimate sigma^2 using sigma(be)^2=RSS/(n-2) Then Beta(be)-Beta / se[Beta(be)] = Beta(be)-Beta / [sigma(be)/sq(Sxx)] Has "Students" t-distribution with n-2 df Where: se = standard error For t-test: x1(avg)-x2(avg)/se(x)[pooled] (Q1) is [sigma(be)/sq(Sxx)] the standard error or the standard deviation? (Q2) if [sigma(be)/sq(Sxx)] is the standard deviation then the standard error should be [sigma(be)/sq(Sxx)]/sqrt(1/n)? Any comment would be greatly appreciated. Roy Carambula . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
