Point estimation

For small datasets, it is easy to understand the where the middle lies. For example, for a dataset that consists of 1, 2, 3, we can see the "middle" is 2. What about with larger datasets where it's not possible to easily guess?

For that we may need the help of a computer or calculator. Let's take a real dataset as an example: consider the Open Intro county dataset. What if we wanted to compute the mean for the pop2000 column which is the estimated population for a county in the United States in the year 2000. Would we want to tally 3,142 counties by hand? Not me. That method is prone to mistakes. I would rather let a computer do the work for me.

So let's go to the Picostat Open Intro county dataset listed above and choose Numerical Summaries from the Analysis dropdown menu as seen here.

Picostat Numerical Summaries Image

A dialog box will appear which allows us to select different options. For more details about the checkboxes, please refer to the Picostat Manual. Here, we will not select any of them for simplicity but the dialog box that appears is shown below.

Picostat Statistics Dialog Box

We can now select a column. With the mouse, click once on the pop2000. You will see the column highlighted like this.

Picostat statistics highlighted column

So now we are ready to do our computation. We can click "Submit" in the dialog box above, we should get an answer back like this:

Picostat Statistics Numerical Summary Image

We can see the average county population in 2000 is 89,650. Curious what the largest and smallest counties are? You can filter the data like this:

Picostat Statistics Data Filter Loving County TX

Do you know what the largest county is?