# R Code for Lecture 6 - Random Variables
# Version 4: Jan 26, 2015
# Cumulative probabilities from Normal distribution
pnorm(175,mean=180,sd=5) - pnorm(170,mean=180,sd=5)
# numerical integration
f <- function(x) cfw_dnorm(x,mean=180,sd=5)
integrate(f, l
# R Code for Lecture 4 - Linear Regression
# Version 4: January 20, 2015
# Least-squares example
fgt = getdata("fidget.txt")
attach(fgt) # This allows us to access the data vectors directly
plot(NEA,Fat, xlab="Nonexercise Activity", ylab="Fat Gain", cex.
# R Code for Lecture 3 - Data Relationships: Correlation
# Version 5: January 14, 2015
# This should be working for you by now, but just in case .
# First, a handy function that you could define and source each time you start R
# This will take a text arg
# R Code for Lecture 2 - Distributions and Graphing
# Version 5: January 12, 2015
# Make Data Set
ccdat = c(17,24,31,18,10)
names(ccdat) = c("G0","G1","S","G2","M")
# Bar Plots
barplot(ccdat) #plain, gray plot
barplot(ccdat, col=rainbow(length(ccdat) #bet
BME 503, Winter 2015
Noll
Syllabus
BME 503 Statistical Methods for BME Winter 2015
Course Description:
This course will cover descriptive statistics, probability theory, distributions for
discrete and continuous variables, hypothesis testing, and analysis
Random Variables
Lecture 6
Comments on HW Solutions
Please show work.
We have asked for the R code to be uploaded, but
that is for verification if we are uncertain of some
aspect of your answer.
Wed like to grade the assignment without looking
at the R co
Inference for
Distributions
Lecture 12
Announcements
HW #5 due Thurs, 2/16
Project #1 due Thurs 2/23
Midterm on 2/23, 7-9pm in IOE 1610
Everything up to and including Thursdays class.
No R, but will need calculator.
Review Tues 2/21. No regular class on 2
Data Relationships:
Correlation
Lecture 3
HW #1
Due 1 week from today at the end of class.
Must use R for all problems (though some of them
could be done by hand & Z-table)
Hand in hard copy in class (code not necessary)
Upload R code to CTools
Other matt
Assignment 3 BME 503
Due Thursday February 5, 2015
General guidelines:
Show all of your work and do things in R, where possible. Some of this assignment will required some
paper and pencil work.
Label all axes and title your graphs (using R). Include un
Data Distributions
and Graphing
Lecture 2
Announcements
Lectures are now automatically recorded and will be
posted on the CTools site. Please let me know if you
have problems with this.
The class now officially has more enrollees than seats
in this room w
Yi-Ren Wang
UMID: 51693360
Assignment 2 BME 503
1a
There are four distinct clusters of data points. This is indeed the case as the amount of substance in ng is one of
four possible values, 0.25, 1.00, 5.00, or 20.00. In addition, the corresponding gas chr
1a The populations in NewYork is the most.
1b The open space in New York is the most too.
1d
1e
1f
1e makes it easier to compare cities that seem to have very similar rate values but that can be hard to compare
when not side-by-side. It is also overall ea
Assignment 1 BME 503
Due Thursday, January 22, 2015
General guidelines:
You can get the data files in R using the getdata command we discussed in class.
Show all of your work and do everything in R, keeping all significant figures until you get to your
Yi-Ren Wang
UMID: 51693360
Assignment 3 BME 503
1a
1b
Area under the density curve is
1c
Calculate the area when Y is between 0.75 and 1.25: (0.5*0.75) + (0.5*0.25/2) = 0.4375
P = 43.75%
2
3a
1. f(y) 0
2. The area under the curve f(y) is 1
3b
3c
P(0 Y 1 m
Assignment 2 BME 503
Due Thursday, January 29, 2015
General guidelines:
You can get the data files in R using the getdata command we discussed in class.
Show all of your work and do everything in R, keeping all significant figures until your final answe
Data Relationships:
Linear Regression
Lecture 4
Preview
Last time: we talked about how to use R
Correlations between two data sets
2 (or more) measurements on each data type
Linear relationship
Both measurements contain noise
Today: regression
One variabl
Probability
Lecture 5
Randomness
Probability has its roots in modeling games of chance
until it was applied to other physical problems in the
18th century
We call a phenomenon random if individual
outcomes are uncertain but there is nonetheless a
regular
Two Sample Statistics
Lecture 14
Two-Sample Problems
We commonly want to compare the responses of two
groups
How do low/high calcium levels affect blood
pressure?
How is tumor growth affected by a drug/placebo?
How does matrix stiffness influence cell
pro
Comparing Two
Proportions
Lecture 16
Comparing Proportions
Often we want to compare the proportions between
two groups, Population 1 and Population 2, and their
individual proportions of successes p1 and p2
We will use the following notation
Population
Sa