[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 
GraphicalAnalysis.

I would need to be able to do the following:

    produce a graphical depiction of :
       Oneway ANOVA
       Standard Deviation (Lines :JMPin)
       Means(Diamonds, Dots :JMPin)
       Data Points
       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>
    RSquare
    RSquare Adj
    Root Mean Square Error
    Mean of Response
    Observations (or Sum Wgts)
    <Analysis of Variance>
    Source   DF   Sum of Squares   Mean Square   F-Ratio
    Model
    Error
    C-Total
    <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 
information:

    A graph that would have points depicting the data entered with  a 
"Fit Line"
    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 
the graph...
    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 mailing list