What is the skewness index, sk, for the distribution of cholesterol levels in the Chapman study?

What is the skewness index, sk, for the distribution of...

Key Concepts

Skewness

Skewness is a statistical measure that quantifies the asymmetry of a probability distribution around its mean. It indicates whether the data tends to have a longer tail on one side (positive skewness for a longer right tail or negative skewness for a longer left tail), which is important for understanding the behavior of the distribution.

Third Central Moment

The calculation of skewness relies on the third central moment, which involves taking the average of the cubed deviations from the mean. This moment captures both the magnitude and the direction of the deviations, providing insight into the asymmetry of the data distribution.

Normalization by Standard Deviation

In the skewness formula, the third central moment is normalized by the cube of the standard deviation. This step ensures that the skewness index is dimensionless and allows for meaningful comparisons across different datasets or variables regardless of their scales.

Descriptive Statistics

Skewness is a part of descriptive statistics, which summarizes the main features of a dataset. It complements other measures like mean and variance by providing additional information about the shape of the data distribution, which is crucial for proper data interpretation and subsequent analysis.

Transcript

00:01 So this problem has a lot of answers because there's a lot of deciles.

00:04 But we'll start by showing you the longer way, all the work for the 10th and 20th, and then we'll kind of go from there.

00:10 So the z -score formula, x minus mean over standard deviation, and the mean is 55 and the standard deviation is 15 .5.

00:18 So if i want the first decile, the 10 % there, i need the z -score for that spot.

00:24 And so on the calculator, there's a button called inverse normal.

00:27 If it's a ti -84, you hit second or shift, vars, v -a -r -s, and it's the third one down.

00:33 And you put in 0 .10 and that'll give you a z -score of negative 1 .28.

00:40 And so i take that z -score and i put it in the formula for where the z is, negative 1 .28.

00:47 And then i'm going to solve for x.

00:50 X is missing minus the mean over 15 .5.

00:53 To solve for x, i'm going to do some algebra.

00:55 First of all, i'm going to multiply 15 .5 times negative 1 .28.

01:03 And then i'm going to add 55.

01:07 And when i do that, i get that first decile at 35 .16.

01:14 So there's your first one, 0 .16.

01:16 If i do the same process for the 20 percentile there, inverse normal of 0 .20 is negative 0 .84, negative z there.

01:30 Then i'm plugging it in here for z, negative 0 .84.

01:34 And then solving for x.

01:37 So i'm going to times by 15 .5 and add 55 again.

01:41 And that gives me 41 .98.

01:48 That's the 20th.

01:50 And then without the drawings, but if you go to the 30th, if you do inverse normal of 0 .30, you get this z of negative 0 .52.

01:59 I'm going to solve for x by multiplying these two numbers and then adding the 55, just like i was doing the other ones.

02:06 So the 30th, that decile is 46 .94.

02:15 Same process for all the rest.

02:17 The 40th one, inverse normal 0 .40 is negative 0 .25.

02:21 Solving for x by times and then add.

02:24 That 40th one is 51 .13.

02:30 The nice thing about the 50th one is that is just the mean...

Please give Ace some feedback