Can we say Standard error of estimate (Sxy) is a standard deviation of the error terms in regression model?
2007-05-05 14:43:59 · 2 answers · asked by chiwee_tan 1 in Science & Mathematics ➔ Mathematics
Maybe I don't understand your notation. Usually Sxy denotes the sample covariance between x and y.
The standard error of what estimate? I don't have any idea.
At any rate, the standard deviation of the error is not a standard error. It is measuring how far the y values are from the regression line.
2007-05-05 14:58:05 · answer #1 · answered by blahb31 6 · 0⤊ 0⤋
We may not have the same understanding of what Sxy is, when you do a regression, you operate on reducing the variance of the observations. In an ANOVA, one df can be removed to explain the effect of the regression. Think of this as tilting an horizontal line that passes through y = mean, and taking variances from each datum to the tilted line. The tilted line that gives you the least sums of variances is the regression line. The sums THESE variances is the residual, and the error term. So I think you would be in error.
2007-05-05 22:02:46 · answer #2 · answered by cattbarf 7 · 0⤊ 0⤋