ACF function confidence intervals

[marymil]

I'm using the acf function with the modifier type = 'covariance'

When I use the modifier, it gets rid of the confidence intervals. How can I add them back into the acf function while maintaining the covariance modifier?

Thanks!

[mfherman]

Welcome, @marymil! It is going to be easier for folks around here to help if you include a sample of your data as well as the code you've used to get to where you are stuck. Without knowing the structure of the data or what you've already tried, it's very hard to help.

See this post for more information about best practices for writing questions and including a reproducible example (reprex).

FAQ: Tips for writing R-related questions

First and foremost, please ask! A core goal of community is to be a friendly place to chat about topics related to data science, R, and RStudio. We know that posting to technical forums can be intimidating. But know that many here would love to see you overcome your inhibition and engage with us. Here are a few tips some folks here think might be helpful. Before you post Check Out R Documentation - R has built in documentation on packages and functions. For example typing ?lm into your R console will open the documentation on the lm function. Search - Be sure to search for the basic keywords of your question with your favorite search engine, Stack Overflow, and R-Help. Stack Overflo…

[aosmith]

I looked through some of the code for plot.acf() and if I'm reading things correctly I don't think the function makes confidence intervals for covariances.

To look at the underlying code I ran stats:::plot.acf in my Console. This is the function that plots the output of acf(), and it is this function that controls the creation and display of any confidence intervals.

Here are the lines that I thought were relevant.
First, a logical vector based on what type of acf being computed (which is FALSE if the type is covariance):

with.ci <- ci < 0 && x$type != "covariance"

Then the calculation of initial confidence interval limits based on the logical vector above. Looks like if the result is FALSE this returns c(0, 0), which is why I think you don't get CI for covariances.

clim0 <- if (with.ci) 
          qnorm((1 + ci)/2)/sqrt(x$n.used)
         else c(0, 0)

There is a lot more code in the function so I could have missed something, but I didn't see anything else where the function built CI for covariances.

If you know how to calculate a confidence interval for a covariance you could potentially add these on to the plot via lines(), since plot.acf is built on base R plotting functions.