R help - Jun 2010 - BBH2 and FrF2 packages

Hi R HELP,

I consider the 2^3 factorial experiment described at page 177 of
the book Statistics for Experimenters: Design, Innovation, and Discovery
by George E. P. Box, J. Stuart Hunter, William G. Hunter (BHH2).

This example use the following data in file BHH2-Data/tab0502.dat
at ftp://ftp.wiley.com/
in /sci_tech_med/statistics_experimenters/BHH2-Data.zip

  run  T  C  K  y
1   1 -1 -1 -1 60
2   2  1 -1 -1 72
3   3 -1  1 -1 54
4   4  1  1 -1 68
5   5 -1 -1  1 52
6   6  1 -1  1 83
7   7 -1  1  1 45
8   8  1  1  1 80

Using these data and the R BHH2 package, I was not able to reproduce the very
simple results in the BHH2 book.
In particular, the following solution will have no meaning since K is
categorical:

( plan <- lm(y ~ (T+C+K)^2, data = DATA) )
MEPlot(plan) # Main Effects
IAPlot(plan) # Interactions Effects
DanielPlot(plan)
cubePlot(plan, "T", "C", "K")

I decided to rebuilt the data using:

plan <- FrF2(8, 3, factor.names=c("T","C","K"),
default.level=c("-","+"), randomize = FALSE)
( plan <- add.response(plan, y) )

giving:

  T C K  y
1 - - - 60
2 + - - 72
3 - + - 54
4 + + - 68
5 - - + 52
6 + - + 83
7 - + + 45
8 + + + 80
class=design, type= full factorial 

Unfortunately the following plot commands do not work:

MEPlot(plan)
IAPlot(plan)
DanielPlot(plan)

The error is:
Error in MEPlot.design(plan) : 
  The design obj must be of a type containing FrF2 or pb.

Why?

If I add a fake factor to the plan the plot commands work, but the solution will
have no meaning:

plan <- FrF2(8, 4,
factor.names=c("T","C","K","Q"),
default.level=c("-","+"), randomize = FALSE)
( plan <- add.response(plan, y) )
MEPlot(plan)
IAPlot(plan)
DanielPlot(plan)

Sincerely, Andrea B.





	[[alternative HTML version deleted]]

Hi:

MEPlot, IAPlot and cubePlot come from the FrF2 package; the DanielPlot
function is in both package BsMD
and FrF2. Try

library(FrF2)

and then run your code again; it worked for me...

If you check the list of functions in BHH2 under HTML help, you'll find that
none of the plot functions you used below are found in that package, but
they are all found under FrF2.

HTH,
Dennis

On Thu, Jun 24, 2010 at 4:17 PM, Andrea Bernasconi DG <
andrea.bernasconi.dg@gmail.com> wrote:
> Hi R HELP,
>
> I consider the 2^3 factorial experiment described at page 177 of
> the book Statistics for Experimenters: Design, Innovation, and Discovery
> by George E. P. Box, J. Stuart Hunter, William G. Hunter (BHH2).
>
> This example use the following data in file BHH2-Data/tab0502.dat
> at ftp://ftp.wiley.com/
> in /sci_tech_med/statistics_experimenters/BHH2-Data.zip
>
>  run  T  C  K  y
> 1   1 -1 -1 -1 60
> 2   2  1 -1 -1 72
> 3   3 -1  1 -1 54
> 4   4  1  1 -1 68
> 5   5 -1 -1  1 52
> 6   6  1 -1  1 83
> 7   7 -1  1  1 45
> 8   8  1  1  1 80
>
> Using these data and the R BHH2 package, I was not able to reproduce the
> very simple results in the BHH2 book.
> In particular, the following solution will have no meaning since K is
> categorical:
>
> ( plan <- lm(y ~ (T+C+K)^2, data = DATA) )
> MEPlot(plan) # Main Effects
> IAPlot(plan) # Interactions Effects
> DanielPlot(plan)
> cubePlot(plan, "T", "C", "K")
>
> I decided to rebuilt the data using:
>
> plan <- FrF2(8, 3,
factor.names=c("T","C","K"),
default.level=c("-","+"),
> randomize = FALSE)
> ( plan <- add.response(plan, y) )
>
> giving:
>
>  T C K  y
> 1 - - - 60
> 2 + - - 72
> 3 - + - 54
> 4 + + - 68
> 5 - - + 52
> 6 + - + 83
> 7 - + + 45
> 8 + + + 80
> class=design, type= full factorial
>
> Unfortunately the following plot commands do not work:
>
> MEPlot(plan)
> IAPlot(plan)
> DanielPlot(plan)
>
> The error is:
> Error in MEPlot.design(plan) :
>  The design obj must be of a type containing FrF2 or pb.
>
> Why?
>
> If I add a fake factor to the plan the plot commands work, but the solution
> will have no meaning:
>
> plan <- FrF2(8, 4,
factor.names=c("T","C","K","Q"),
> default.level=c("-","+"), randomize = FALSE)
> ( plan <- add.response(plan, y) )
> MEPlot(plan)
> IAPlot(plan)
> DanielPlot(plan)
>
> Sincerely, Andrea B.
>
>
>
>
>
>        [[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]]