piatok 21. novembra 2008

Comparing multiple proportions: The Marascuillo procedure

Comparing multiple proportions: The Marascuillo procedure





from NIST Handbook. includes syntax for calculation








BIOS nice overview of proportions / basics - equivalence of Pearson chi-square test and z-test





Macro from Minitab


This macro performs a multiple comparisons test on proportion data using # Tukey's honest significant difference test.



statistics for lawyers

google books

http://books.google.com/books?id=_bR9Q-EERlIC&pg=PA200&lpg=PA200&dq=comparing+multiple+proportions&source=bl&ots=PTb87Wa-gr&sig=PuEEGFcVwPlqSnhLUF6zXoXJ5p0&hl=en&sa=X&oi=book_result&resnum=5&ct=result#PPA201,M1

pondelok 10. novembra 2008

lognormal distribution - on different methods of calculation of confidence intervals




utorok 7. októbra 2008

SEQUENCING low price, individual genome

Low-Price Mapping SEQUENCING

companies developing/offering genome sequencing

- COMPLETE GENOMICS - 5000 $ /Reid, Radoje Drmanac
- applied biosystems
- Knome - cca 20 customers
- cf Systems Biology Institute as customer for complete genomics

sobota 20. septembra 2008

datamining

datamining in genetic/genomic research

finland Helsinki
DM in LD mapping ... haplotype association etc. PDF

PUBLICATIONS BY HANNU TOIVONEN, UNI helsinki

štvrtok 18. septembra 2008

datamining intro

datamining course materials from Australian national Uni,
- same as links below.
COURSE SLIDES -FROM slideshare.com

by prof Lanzi, some slides look identical to slides provided by ANU

- but generally looks good, COMPREHENSIBLE .....



DATAMINING, UNSUPERVISED RECORD LINKAGE.

Markus Hegland , australia

CHRISTEN
DATAMINING. CHALLENGES, MODELS, METHODS AND ALGORITHMS
year 2003
intro -- mainly from the standpoint of computation science.. algorithms
--
progream - FEBRL year 2008
Febrl - A Freely Available Record Linkage System with a Graphical User Interface Peter Christen Proceedings of the Australasian Workshop on Health Data and Knowledge Management (HDKM), Wollongong, January 2008.

- DISTANCE - EUCLIDEAN, PYTHAGOREAN ETC WOLFRAM MATH



ASSOCIATION RULES

- Support, confidence

Support gives total number of transaction of any particular item are occurring in datasets while confidence gives strength of a data in a dataset, we can say support is probability of A and B while confidence is conditional probability. Association rule based on these two characteristics.


pondelok 15. septembra 2008

c-statistic

____?????? I would like to know whether we can calculate C-statistic using SPSS >> 13.

> If by the "c-statisitc" you are referring to the measure of the > discriminative power of the logistic equation, you can calculate it by > saving the predictive probabilities from the logistic regression > analysis and running a ROC curve with the preditive probabilities as > the "test variable." The c-statistic is the area under the curve > value.


HTH

In R/S-Plus you can just use the lrm function in the Design package or:

mean.rank <- mean(rank(x)[y == 1]) c.index <- (mean.rank - (n1 + 1)/2)/(n - n1)
(where n1 is the number of observations with y=1)
or use the somers2 function in the Hmisc package.

Frank Harrell