Contrast variable: Difference between revisions

Revision as of 16:23, 21 July 2011

Revision as of 02:15, 22 July 2011

Line 1:

AIn [[statistics]], a '''contrast variable''' is defined as a [[linear combination]] of [[random variablesvariable]]s in which the sum of the coefficients is zero<ref name=ZhangPharmacogenomics2009>{{cite journal |author=Zhang XHD

(with each variable representing random values in one of multiple groups involved in a comparison) instead of group means, with the restriction of the coefficients in the combination summing to zero.<ref name=ZhangPharmacogenomics2009>{{cite journal |author=Zhang XHD

|title= A method for effectively comparing gene effects in multiple conditions in RNAi and expression-profiling research

|journal=Pharmacogenomics |volume=10 |issue= |pages=345-58345–58

|year=2009 |month= |pmid= |doi=10.2217/14622416.10.3.345 |url=}}</ref>. Each variable may represent random values in one of multiple groups involved in a comparison. Associated with a contrast variable are two terms,: the standardized mean of contrast variable ([[SMCV]]) and [[c+-probability]]. The SMCV is the ratio of [[mean]] to [[standard deviation]] of a contrast variable and the c+-probability is the probability that a contrast variable obtains a positive value.

contrast variable ([[SMCV]]) and [[c+-probability]]. The SMCV is the ratio of mean to standard deviation of a contrast variable and the c+-probability is the probability that a contrast variable obtains a positive value.

Traditional [[contrast]] is a statistical parameter defined only on the group [[mean]]s. A contrast variable allows us to consider not only group [[mean]]s but also group [[variance]]s in a comparison. In addition, the concept of contrast variable can help to derive the [[effect size]]s across any number of groups readily and smoothly.<ref name=ZhangJBiometBiostat2010>{{cite journal |author=Zhang XHD

Line 13 ⟶ 10:

|url= http://www.omicsonline.org/2155-6180/2155-6180-1-108.php}}</ref> The concepts of contrast variable, [[SMCV]] and [[c+-probability]] were recently proposed for one-way [[ANOVA]] cases <ref name="ZhangPharmacogenomics2009"/> and were then extended to multi-factor [[ANOVA]] cases <ref name=ZhangPharmacogenomics2010>{{cite journal |author=Zhang XHD

|title= Assessing the size of gene or RNAi effects in multifactor high-throughput experiments

|journal=Pharmacogenomics |volume=11 |issue= |pages=199-213199–213

|year=2010 |month= |pmid= |doi=10.2217/PGS.09.136 |url=}}</ref>

<ref name="ZhangJBiometBiostat2010"/>

Line 31 ⟶ 28:

* [[Contrast (statistics)]]

* [[Dual-flashlight plot]]

==References==