Biostatistics 615/815
Statistical Computing
Gonalo Abecasis
Course Objective
Introduce skills required for executing
statistical computing projects
Applications and examples mostly in C.
Can be easily translated into R, etc.
But the focus is on an algori
Cell Communication Project Packet Name
Projects that are not typed and spellchecked (over 10 spelling errors will prove to me that you have not spellchecked)
will receive a 10 point deduction.
You will need a good internet connection to complete much of t
Biostatistics 615/815
Problem Set 10
Due Monday, December 6, 2004
1. For this exercise, you will need the data set ModelFittingData.txt available at
www.sph.umich.edu/csg/abecasis/class/ (Right click on the data link and select
Save as to download). The d
Biostatistics 615/815
Problem Set 9
Due Friday, December 3, 2004
Consider the vector norm function:
f ( x) = x =
x
2
i
i
The function has minimum 0 when x = 0.
We will use this function to examine the efficacy of the Nelder-Mead algorithm for
function min
Biostatistics 615/815
Problem Set 8
Due November 24, 2004
1. Consider the following set of 20 observations drawn from a mixture of two
normal distributions.
-2.876
-2.527
-1.213
-1.111
-1.034
-0.877
-0.645
0.151
0.246
0.409
0.728
0.737
0.819
0.998
1.602
1
Biostatistics 615 / 815
Problem Set 7
Due November 17, 2004
Random Number Generation
1. An old text book on computer simulation recommends the following sequence of
pseudo-random numbers:
Ij+1 = a Ij mod m, with a = 20,403 and m = 215.
What is the period
Biostatistics 615
Problem Set 4
Due October 20, 2004
Elementary Sorts
1. Which of the three elementary methods (selection, insertion and bubble sort) do
you expect to run fastest for an array in reverse-order? Why?
(Hint: You can solve this problem by com
Biostatistics 615/815
Problem Set 2
Due Wednesday, September 29
Floating Point Precision
1. This problem aims to explore the accuracy of floating point representations in
your system. You should investigate the float and double data types in C as well
as
Name: _
Major: _
1. Are you taking this course for credit?
If not, are you planning to attempt the homework sets?
2. What are you hoping to learn from this course?
3. How do you expect to apply statistical computing tools in your future career?
Do you exp
Introduction to Numerical
Optimization
Biostatistics 615/815
Lecture 14
14
Course is More Than Half Done!
If you have comments
they are very welcome
Lectures
Lecture notes
Weekly Homework
Midterm
Content
Last Lecture
Computer generated random number
Merge
Merge Sort
Biostatistics 615/815
Lecture 8
Notes on Problem Set 2
Union Find algorithms
Dynamic Programming
Results were very positive!
You should be gradually becoming
comfortable compiling, debugging and
executing C code
Question 1
How many random
Quick
Quick Sort
Biostatistics 615/815
Lecture 7
Last Lecture: Shell Sort
Gradually bring order to array by:
Sorting sub-arrays including every kth element
Using a series of step sizes k, ending with k = 1
Each pass handles nearly sorted arrays where
Each
Shell
Shell Sort
Biostatistics 615/815
Lecture 6
Housekeeping
Housekeeping Note:
Homework Grading
Weihua Guan is the GSI
He requests that you e-mail him source
code for your assignments to:
code for your assignments to:
[email protected]
Thanks!
Last Lectur
Recursion and
Dynamic Programming
Biostatistics 615/815
Lecture 4
Last Lecture
Principles for analysis of algorithms
Empirical Analysis
Theoretical Analysis
Common relationships between inputs
and running time
Described two simple search algorithms
Recu
Fall 2015 Biostat 615 Homework #1 (Total 40 pts)
Due by Wednesday September 30th, 2015 noon as a compressed le hw1.tar.gz containing the
required C+ source code les for the following four problems. Please read the instruction carefully
before you start to