R help - May 2013 - Pooled mean square error for regression

Hi,

I would like to compare two batches of a product over time to see if they behave
similarly, i.e same slope and intercept. This is a stability study. I will be
performing ANCOVA analysis for this. According to ICH and FDA guidelines, if the
test rejects the hypothesis of equality of slopes, then I should fit a
regression line to each batch where the intercept and the slope are the
individual ones but I should use the pooled mean square error calculated from
all batches.

I wonder if anyone could help me to feed the calculated pooled mean square error
in the function lm() or if there is another way to perform such analysis.

Thanks for your help,

Dave
 		 	   		  
	[[alternative HTML version deleted]]

I'm not sure I understand.  If you fit an ANCOVA using the lm() function,
like this,
    fit <- lm(Y ~ X*as.factor(batch))
    summary(aov(fit))
and the hypothesis for equality of slopes is rejected, i.e., the
X:as.factor(batch)) term is significant, then this same model already has
batch-specific intercepts and slopes and a pooled mean squared error.  What
more do you need?

Jean


On Fri, May 31, 2013 at 5:41 AM, David A. <dasolexa@hotmail.com> wrote:
> Hi,
>
> I would like to compare two batches of a product over time to see if they
> behave similarly, i.e same slope and intercept. This is a stability study.
> I will be performing ANCOVA analysis for this. According to ICH and FDA
> guidelines, if the test rejects the hypothesis of equality of slopes, then
> I should fit a regression line to each batch where the intercept and the
> slope are the individual ones but I should use the pooled mean square error
> calculated from all batches.
>
> I wonder if anyone could help me to feed the calculated pooled mean square
> error in the function lm() or if there is another way to perform such
> analysis.
>
> Thanks for your help,
>
> Dave
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]