z.test returning NA's

[gdunster]

Hello,

I am using the z.test function from the BSDA package in an attempt to analyze some data and I keep getting an output of "NA". There are no errors when I run the test, but all of my outputs ( z=, p-value =, etc) simply say NA. Below is the code I have written to run the analysis, hopefully you can provide some insight.

Dat1 <- Dat[Dat$Year == 1,]
Dat2 <- Dat[Dat$Year == 2,]

sd16 <- sd(Dat1$ColumnA, na.rm = T)
sd17 <- sd(Dat2$ColumnA, na.rm = T)

z.test(x = Dat1$ColumnA, sigma.x = sd16, y = Dat2$ColumnA, sigma.y = sd17)

[alanarnholt]

The function does not automatically remove NA values. I do not have your data but I suspect you have NA values in one or both of Dat1$ColumnA and Dat2$ColumnA. Try

`
Dat1 <- Dat[Dat$Year == 1,]
Dat2 <- Dat[Dat$Year == 2,]

sd16 <- sd(Dat1$ColumnA, na.rm = T)
sd17 <- sd(Dat2$ColumnA, na.rm = T)

z.test(x = na.omit(Dat1$ColumnA), sigma.x = sd16, y = na.omit(Dat2$ColumnA), sigma.y = sd17)
`
let me know if that solves your problem. I will try to write some code in the future to do this automatically.

[alanarnholt]

Thanks for bringing this to my attention. I just updated the code to remove the NAs as I claimed in the documentation. Please use the development version 1.2.1 for the latest fix. I will update BSDA on CRAN in a month or so. Let me know if the code in 1.2.1 solves your issue.

[gdunster]

It worked perfectly! Thank you so much for your help!

[alanarnholt]

Issue closed