[R] Newbie Help:: Honors Chem. Project:: How to use R instead of JMP IN and GraphicalAnalysis...
Aaron "Snail:Lewis" Dinkin
snail at nycap.rr.com
Sun Nov 16 06:04:53 CET 2003
Greetings, and Salutations to the wonderful people of the R-Project!!!
I have a question:
I use JMP IN and GraphicalAnalysis for my Honors Chem. Projects. I
would like to be pointed in the right direction: I would like to know
how to use R to obtain the same data I would if I were to use JMP IN and
I would need to be able to do the following:
produce a graphical depiction of :
Standard Deviation (Lines :JMPin)
Means(Diamonds, Dots :JMPin)
Standard Error (Bars/Lines :JMPin)
I would have to be able to "Fit X by Y" and make the X-axis Proportional
I would need to obtain the following numerical data:
<Summary of Fit>
Root Mean Square Error
Mean of Response
Observations (or Sum Wgts)
<Analysis of Variance>
Source DF Sum of Squares Mean Square F-Ratio
<Means for Oneway ANOVA>
Level Number Mean Std-Error
<Means and Std Deviations>
Level Number Mean Std-Dev Std-Err-Mean
I would also need to produce data that would illustrate my precision of
measurements and my accuracy of measurements... Either Graphically or
Numerically or both...
My class uses %CV to illustrate precision where %CV= (sd()/mean)*100
But any or all other methods used to calculate precision (and accuracy)
would be fine too...
<The above was to describe a replacement for JMP IN>
For a replacement for GraphicalAnalysis, I would need the following
A graph that would have points depicting the data entered with a
a "Curve Fit" that would be set for "Linear" a Linear Equation...
The "Liner Fit" would be depicted on the graph, the equation of the
line in y = mx+b format, the m(slope) of the line, the b(Y-intercept)
of the line, and the Correlation [coefficient] of the line...
I would have to be able to annotate the graph with my name, a title
for the graph, the date, and the equation of the line...
I would also need to be able to print out the dataset used to create
also, I would need interpolative functionality: I would need to be
able to figure out data while having only one known variable (ie.: we
had to analyze the concentration of a pigment using a Spectrometer we
had two variables: Concentration and Absorbance, several of the samples
had "Unknown" concentrations while having known absorbances, we needed
to find their Concentrations by interpolating the "Linear Fit For:
Spectrometer Data" which entailed finding the spot where Absorbance
(foo) hit the line, then we would assign the (bar) value (x axis value
to y value) to Concentration, the Unknowns are now known!!!
Many Thanks In Advance to the Kind peoples whom helped me find, and
understand the information I requested!!!
More information about the R-help