On the error bar and statistical significance

Take two random series of 20 values. What you can tell about their difference according to the visualization of their confidence interval: nearly nothing !

Let use this little script:
library(HelpersMG)

x <- rnorm(20, mean=11.8, sd=2)
y <- rnorm(20, mean=10, sd=2)

t <- t.test(x, y, var.equal = TRUE)
w <- series.compare(x, y, criterion = c("BIC"), var.equal = TRUE)

plot_errbar(x=1:2, y=c(mean(x), mean(y)), errbar.y=1.96*c(sd(x), sd(y)),
            las=1, bty="n", xlab="", ylab="", ylim=c(0, 20), xlim=c(0, 3))
plot_errbar(x=(1:2)+0.1, y=c(mean(x), mean(y)), errbar.y=2*c(sd(x)/sqrt(20), sd(y)/sqrt(20)),
            las=1, bty="n", xlab="", ylab="", ylim=c(0, 20), xlim=c(0, 3), add=TRUE, col="red",
            errbar.col = "red")

text(x = 1.5, y=2, labels = paste0("p = ", format(t$p.value, digits = 5)))
text(x = 1.5, y=3, labels = paste0("w = ", format(w["BICw(identical)"], digits = 5)))

legend("topright", legend=c("Standard deviation", "Standard error"), col=c("black", "red"), lty=1)


The result is this one:


First, a common mistake is to look at the dispersion of the values measured by standard deviation (in black) and then conclude about the difference or not of the mean. Clearly here there is a large overlap of standard deviation but the standard errors (i.e. how is known the average) do not overlap. This is reflected by the low p value= 0.003. But always remember that the p value is not the probability that the two series are different ! The probability that the two series are different is better measured by the w-value (Girondot and Guillon, 2018).
If you claim that the means are different and you believe that the probability that you are wrong is p = 0.003,  in reality you are wrong with a probability of 0.10 !

In conclusion:
1/ do not use p-value
2/ do not use standard deviation dispersion bars to conclude about difference of means
3/ use w-value !

Girondot, M. & Guillon, J.-M. 2018. The w-value: An alternative to t- and X2 tests. Journal of Biostatistics & Biometrics, 1, 1-4.


Commentaires

Enregistrer un commentaire

Posts les plus consultés de ce blog

Standard error from Hessian Matrix... what can be done when problem occurs

An Hessian matrix is a square matrix of partial second order derivative. From the square-root of the inverse of the diagonal, it is possible to estimate the standard error of parameters. Let take an example: val=rnorm(100, mean=20, sd=5) # Return -ln L of values in val in Gaussian distribution with mean and sd in par fitnorm<-function(par, val) {   -sum(dnorm(val, par["mean"], par["sd"], log = TRUE)) } # Initial values for search p<-c(mean=20, sd=5) # fit the model result <- optim(par=p, fn=fitnorm, val=val, method="BFGS", hessian=TRUE) # Inverse the hessian matrix to get SE for each parameters mathessian <- result$hessian inversemathessian <- solve(mathessian) res <- sqrt(diag(inversemathessian)) # results data.frame(Mean=result$par, SE=res)   library(MASS) (r<-fitdistr(val, "normal")) It works well. However, the inverse of the matrix cannot be calculated if the second order derivative for some paramete
Lire la suite

stepAIC from package MASS with AICc

Due to a bug in dropterm() and addterm() in package MASS, it is impossible to use AICc with stepAIC() . Here is a solution. Enter these commands and now you can use: stepAIC(g0, criteria="AICc") or stepAIC(g0, criteria="BIC") For example: datax <- data.frame(y=rnorm(100), x1=rnorm(100),                     x2=rnorm(100), x3=rnorm(100), x4=rnorm(100)) g0 <- glm(y ~ x1+x2+x3+x4, data=datax) gx_AICc <- stepAIC(g0, criteria="AICc") gx_AIC <- stepAIC(g0, criteria="AIC") gx_AIC <- stepAIC(g0, criteria="BIC")
Lire la suite

Install treemix in ubuntu 20.04

Informations are available here  https://bitbucket.org/nygcresearch/treemix/downloads/ In terminal, enter for  Ubuntu 20.04: wget https://bitbucket.org/nygcresearch/treemix/downloads/treemix-1.13.tar.gz tar -xvf treemix-1.13.tar.gz sudo apt-get install libgsl-dev sudo apt install libboost-dev sudo apt install libboost-all-dev cd treemix-1.13/ ./configure make sudo make install And it works: $ treemix TreeMix v. 1.13 $Revision: 231 $ Options: -h display this help -i [file name] input file -o [stem] output stem (will be [stem].treeout.gz, [stem].cov.gz, [stem].modelcov.gz) -k [int] number of SNPs per block for estimation of covariance matrix (1) -global Do a round of global rearrangements after adding all populations -tf [file name] Read the tree topology from a file, rather than estimating it -m [int] number of migration edges to add (0) -root [string] comma-delimited list of populations to set on one side of the root (for migration) -g [vertices file name] [edges file name] read the g
Lire la suite