|
Homework PHY 5157 Fall 2005
- For September 1: Prepare to talk 2-5 minutes about your
science background and interest in MCMC methods.
If you have already ideas about your final project,
include them too.
- For September 6: Read chapter 1 up to section 1.4 (included).
a0101_02: Confirm the correct installation of
Marsaglia's random number generator.
- For September 8: Complete your estimate of pi (unless you are
already done).
Fortran program for download: pi_v1.f .
C program for download: pi_v1.c .
- For September 13: Complete the peaked distribution function
gnuplot for assignment a0104_both.
- For September 20: Download and run the program
polls.f ( polls.c
for C users).
The program includes data from an early poll of the 1992 US
presidential election.
With what probability would Clinton win the elections according
to the bootstrap analysis of the program?
E-mail your conclusions (in percent rounded to two digits) to
the instructor.
Compile and run a0108_02: Binning.
Finish reading chapter 1 of the book.
- For September 22: Here are three formatted data files and
programs which reads them:
dat1.d ,
dat2.d ,
dat3.d ,
pread.f ,
pread.c .
Download the data and a read program.
Use steb0.f of ForLib to calculate
mean values and error bars of the data.
E-mail your results.
Perform some Gaussian difference tests interactively and
reproduce assignment a0201_09.
- For September 27: Compile and run a0202_01, a0202_02, a0203_01,
a0203_02, a0204_01, a0205_01, a0205_02.
Q-values of the quiz: diff.txt .
- For September 29: Run the program of assignment a0204_03.
For the second part enlarge the statistics by a factor
of 100 to 1,920,000.
If the program crashes in the second part understand why
and fix it.
E-mail your conclusions.
Compile and run a0206_01, a0206_04.
- For October 6: Bin the data set dat1.d from September 22
into jackknife bins.
Calculate their jackknife error bars with stebj0.f.
Continue with: hw050929.pdf .
- For October 18: Study the pointer examples given in
a0301_01.
Sample energy data for the 2D Ising model: a0301_02.
- For October 20: Complete your worksheet, which we started
in class: metwsheet.txt .
Reading: Chapter 2 up to 2.7. Chapter 3 (omitting 3.2.2 and
3.3.2).
- For November 1: Complete the worksheet
hw051025.pdf .
- For November 3: Complete the worksheet
hw051027.pdf .
Plot from this homework:
acpt_tauint.pdf .
- For November 8: Perform a fit of the form beta_c(L)=beta_c+a/L**3
to the data of the file beta_pc.dat.
The data are pseudo-transition values
for the 3D 3-state
Potts model from the equal heights definition for the
energy
histogram. E-mail your estimate beta_c +/- error bar.
- For November 22: Repeat the fit of a0403_04 for data from a
20*20 lattice
(increase the statistics of the MC run
accordingly). E-mail the estimates of
the leading and sub-leading exponential
autocorrelation times to the instructor.
- For November 29: Create the multicanonical data of a0501_02
for the
for the 2D 10-state Potts model. Use the
data to give an estimate of the
latent heat from this
simulation (compare a0501_05 and _06).
- For December 1: Integrate the folder a0601_01c (which you got
e-mailed as a0601_01.tgz)
into your STMC Fortran package.
Configure your classroom account for the use of MPI
(see
the readme.txt file), if you did not already do so in class.
- For December 8: Updated instructions for implementing MPI:
readme2.txt .
- Final Presentations:
Bazavov_BiasedMC.pdf.
Brolley_Meteorology.ppt.
Fleming_PDEs.ppt.
Ren_RandomNumbers.ppt.
Mersenne Twister random number generator C-Code
rng_afm.c (Jane Ren).
Robin_MD.ppt.
Thompson_Micromagnetical.ppt.
Thompson_Langevin.ppt.
Tran_Quantum-Hall.ppt.
Wierschem_PathIntegralMC.pdf.
Back to MCMC main page.
|