Discover Your Electronic Receipt Sample for R&D to Simplify Your Workflow

Electronic receipt sample for R&D

now we're going to use r as a simulation platform and we're going to try to answer this question how does the sample size affect the sampling distribution of means we're going to use the iris data set and in the IRS data set there is a column pedal length that's the population that we're going to be interested in so we can think of our population as consisting of these 150 values of pedal length so what we want to do is simulate this sampling process which means that we want to iteratively draw samples from this population calculate a sample mean then store that sample mean so let's begin by deciding how many iterations we want maybe we'll start with 10,000 and see how that goes and at every iteration we're going to be drawing a sample from the population and that sample has some sample size so maybe we're drawing five flowers at a time I'm going to initialize a vector of Nas that we can use to store our sample means I'll just call it means and because this is an iterative process I'm going to use a for Loop so I'll say for I and one through iter and then what I want to do is take my sample so I'll store that temporarily in D and I'll just say sample from Iris pedal length and my sample size is n and then once I've got my sample I want to calculate the mean and store it so I'll store that in the I slot in this object means and it'll just be mean d let's take a look at our sampling distribution okay there we go and I want to draw a vertical line at the true population mean which we actually know in this case because we know the entire population so I'll copy that and I'll just say AB line I'm going to make a vertical line the location of this mean and I want it to be dashed I want want it to be a little thicker and I want it to be blue okay our original question is how does sample size affect the sampling distribution of means so here's our sampling distribution and our sample size in this case was five let's see what happens when we vary this sample size I'll change it from 5 to 20 and because I've reused n throughout I could have just written five here but instead I used n so now I just have to change this one value and the rest of my code will work just fine so let's see what happens when we increase sample size to 20 well notice the x axis scale here is quite a bit different so we're going from about 2.5 to five whereas when we had sample size of five we were going from about 1 to six okay what happens if we increase the sample size even more say to 90 well then we've got even tighter bounds right so we're going from about 3.4 to 4.2 with sample sizes of 90 so we've answered our original question and we've done it through simulation of course there are equations that describe the relationship between sample size and the variance in this sampling distribution but being able to actually simulate that process can give you a unique insight into it which is kind of nice