r - extract verbatim p-value from cor.test -
i ran correlation analysis between 2 vectors , need plot -log10(pvalue)...when correlation runs, have following results:
data <- read.table("test-data.txt") <- data[,1] b <- data[,2] test <- cor.test(a, b) test pearson's product-moment correlation data: , b t = 8.3704, df = 10352, p-value < 2.2e-16 alternative hypothesis: true correlation not equal 0 95 percent confidence interval: 0.06282908 0.10109436 sample estimates: cor 0.08199194
but when extract p-value test$p.value
comes out zero..this means when calculate -log10(pv) inf
results...which cannot plot.
however, -log(2.2e-16, base=10)
15.65758
want
so question is: how maintain exponential number?
also, sample of series of 81 correlations i'm running have plotted together, have same problem 30 out of 81
your p-value not 2.2e-16, less that. means computationally zero. can skip zeros, or set 0 arbitrarily small 2.2e-16.
ifelse(pv==0,-log(2.2e-16,base=10),-log(pv,base=10))
of course, 30 out of 81 times not know p-value is, small.
Comments
Post a Comment