Last thing I want to talk about is removing missing values or NA values from an object. This is a very common operation in, in, in data analysis because most realistic data have lots of missing values. And so the way you can do this for, of either a vector, or a matrix, or a data frame is you want to create a logical vector which tells you where the NA's are and so that you can remove them by sub-setting. So here I've got a, a vector x, this is a very simple example. Which has the numbers 1, 2 and 4 and 5. But then there are missing elements NA in the third position and in the fifth position. So what I want to do is I want to get a vector back that's just one two four five, the non missing values. And I want to strip out the missing value so I can maybe do some computation. So, what I, the first thing I do is I use the is.na function to, to go through the vector and tell me which elements are NA and I create a new vector called, which I called bad here. So, bad is going to be a logical vector which tells, which, which is true if the element is missing and false if it's not missing. So, even though I haven't printed it here the, the, the bad vector is going to be a logical vector that has false, false, true, false, true, false. Right, because the third and the fifth elements are missing. So, when I. Now that tells me which ones are missing but actually I don't want the ones that are missing, I want the ones that are non missing. So I need to take the kind of the opposite of bad which I can use with the bang operator or exclamation point. So now I take x single bracket bang bad and that gives me the good elements which are 1, 2, 4 and 5. So what if there are a multiple vectors or multiple objects and you want to take and each one has a kind of missing values in slightly different places and you kind of, you want to take the subset of all the objects that have no missing values, all right? So here I created an x, a vector called x. Which is 1, 2, 4 and 5, and missing values scattered about. And then y is, is a character vector with also some missing values in it. So I can use the complete cases function on both vectors which will give me a vector that tells me which of, of, of the two different vectors, which ones had were, which, which are, which positions are there that have both elements non missing. So you can see the first two are, are both non missing. I got 1, 2, in the first one, and a, b in the second one. The third one's missing, the fourth one is non missing the, the fifth one is missing and the sixth one is non missing. So, of, for the both, for both vectors x and y, I want the first, the second, the fourth, and the sixth elements. So now when I subset x, I get the good elements of that, and when I subset y, I also get the good elements of that. So that's how I can look at multiple objects and kind of subset all the missing values out to get the good elements. You could also remove you can also use complete cases to remove missing values from data frames. So here I've got a simple data frame where I'm showing the first six rows. As you can see there are six columns to this data frame so there's six variables. And there's some missing values in the ozone variable and there's some missing values in the solar.r variable. And so all I want is the is the, is the, are the rows of the data frame where all the values are non missing, right? So in this simple example, the rows that I want are rows 1 through 4. So I can use complete cases on air, on the airquality data frame. And I create a, a logical vector that I called good here, so the logical vector here tells me which rows are complete. And then when I subset out the air quality matrix take and take out the first couple of rows you can see that I now, that none of the rows have any missing values in them. So that's subsetting out missing values. And there, in complete cases of very handy function which is when, when you have multiple sets of vectors or dat, or large data frames or you want to subset all out, all the missing values.
