Mean
Mean, typically denoted by \(\bar{x}\), measures the central value of the data set. It is the sum of all the values divided by the number of values.
\(\bar{x}=\frac{\sum_{i=1}^{n}{x_{i}}}{n}\)
Example
For example, let’s say in a class of 4 children, Alex, Beth, Charlie and Dawn, scored 3, 8, 21 and 9 for a test. As a teacher, you want to find out the average score for the class. We can use the mean formula to find this. In this case, for the sample \(X=\{3,8,21,9\}\), the sample mean is
\(\bar{x}=\frac{\sum_{i=1}^{4}{x_{i}}}{4}=\frac{3+8+21+9}{4}=10.25\)
Code
In code, this can be done as follows.
// create an array of doubles for our dataset
val values = doubleArrayOf(3.0, 8.0, 21.0, 9.0)
// get the array size
val n = values.size
var total = 0.0
var sample_mean : Double
// find the sum of all the values
for (i in 0..n-1) {
total += values.get(i)
}
sample_mean = total / n
println("Sample size: " + n)
println("Sample mean: " + sample_mean)
Sample size: 4
Sample mean: 10.25
In NM Dev, we can simply compute the sample of a dataset using the class Mean without having to come up with the algorithm ourselves.
// create an array of doubles for our dataset
val values = doubleArrayOf(3.0, 8.0, 21.0, 9.0)
// create the Mean object using the array that we created
val mean = Mean(values)
println("Sample size: " + mean.N())
println("Sample mean: " + mean.value())
Sample size: 4
Sample mean: 10.25
Weighted Mean
Weighted mean allows certain values in the dataset to weigh more than others. Each value in the dataset can have its own weight \(w_{i}\). It is the sum of all the values multiplied by each of their weights divided by the sum of all their weights.
\(\bar{x}^{*}=\frac{\sum_{i=1}^{n}{x_{i}w_{i}}}{\sum_{i=1}^{n}{w_{i}}}\)
Example
Let’s say for example, a student took 5 subjects, English, Physics, Math, Economics and Computing and each subject has a different weight when calculating the student’s overall grade. The weights are as follows.
Subject | Weight |
English | 2 |
Physics | 3 |
Math | 4 |
Economics | 1 |
Computing | 5 |
The student’s grades for each subject are as follows.
Subject | Grade |
English | 88 |
Physics | 94 |
Math | 69 |
Economics | 66 |
Computing | 80 |
In order to find the student’s overall grade through the weighted mean formula, we must first multiply the grade for each subject with their respective weights.
Subject | Weighted Grade |
English | 2 * 88 = 176 |
Physics | 3 * 94 = 282 |
Math | 4 * 69 = 276 |
Economics | 1 * 66 = 66 |
Computing | 5 * 80 = 400 |
We can then find the average of the weighted grades by finding the sum of all the weighted grades and then dividing it by the total weight.
Sum of weighted grades\(=176+282+276+66+400=1200\)
Total weight\(=2+3+4+1+5=15\)
Weighted mean\(=1200/15=80\)
Hence, the student’s overall grade is 80.
The above example can be defined mathematically as follows. The student’s grades for each subject are \(X=\{88,94,69,66,80\}\) and each subject weighs differently in calculating the overall grade, \(W=\{2,3,4,1,5\}\), the weighted sample mean is
\(\bar{x}^{*}=\frac{\sum_{i=1}^{5}{x_{i}w_{i}}}{\sum_{i=1}^{5}{w_{i}}}=\frac{88\times{2}+98\times{3}+69\times{4}+66\times{1}+80\times{5}}{2+3+4+1+5}=80\)
Code
Translating this into code, we can use NM Dev’s class WeightedMean.
// create arrays for our dataset and our weights
val values = doubleArrayOf(88.0, 94.0, 69.0, 66.0, 80.0)
val weights = doubleArrayOf(2.0, 3.0, 4.0, 1.0, 5.0)
// create the WeightedMean object
val weighted_mean = WeightedMean(values, weights)
println("Sample weighted mean: " + weighted_mean.value())
Sample weighted mean: 80.0
Thinking Time
Mean is used everywhere in the real world. Some use cases are:
- Finding the average salary in a certain field
- Finding the average expenditures of households
- Finding the average time it takes to run a race
- Finding the amount of hours spent on phones each week
