Assignment 3 due in class on Tuesday, May 9.
Comments on Assignment 2
It may help to print every question
and then answer it so you don't miss any.
I'm looking for complete answers to
the homework questions and a demonstration that you've synthesized class
and book material.
Non-linear least squares:
given starting values, nls proceeds
via iteration to find parameter values that minimize the squared difference.
The solution space has many possible
local minima, so starting seeds are important.
format: nls (
Understanding more complex Splus objects
In the object-oriented framework, objects
are not simply collections of data.
They are particular instances (instantiations)
of particular classes.
Objects include information
(which might consist of data typed to other classes)
Operations defined upon classes.
Example:
> module(spatial)
> class(bramble.spp)
returns the class of bramble. It's
spp - spatial point pattern. More generally it's a data frame.
> help(spp) : will tell you more about
spatial point pattern objects
Consider the K-function object named
k produced via:
> k_Khat(bramble.spp)
k belongs to a class (could also be
in a super/sub class): class(k)
k consists of several "things":
> names(k)
> k
and a variety of functions may be defined
for it
> bramble.spp
> plot(bramble.spp)
Side: what's the difference between
> plot(bramble.spp)
and > plot(bramble.spp$x, bramble.spp$y)
You can define objects to be particular
classes. Try this:
> x_rnorm(50, 10, 3)
# creates 50 random x values
> y_rnorm(50, 10, 4)
# creates 50 random y values
> plot(x,y) # Gee, it looks
like a point pattern...
> k2_Khat(x,y) # did this work?
try this instead:
> k2_Khat(as.spp(as.data.frame(cbind(x,y))))
Working with Linear Models in SPlus
A depressing dataset, and oh so aspatial!
ovarian
Plot futime against age: plot(ovarian$futime,
ovarian$age)
Perform some regressions:
> ovlm_lm(futime ~ age, data=ovarian)
> ovlm2_lm(futime ~ age + residual.dz
+ rx + ecog.ps, data=ovarian)
lm objects are complex: > names(ovlm)
Useful operations on lm objects include
summary and print
You can examine and work with parts
of the "superobject":
> plot(ovlm$resid)
> abline (0,0)
> qqnorm (ovlm$resid)
Using the interp function
Use to generate surfaces from point
data.
> z_rnorm(50, 100, 20)
Back to Main
G274 TA Page