You are on page 1of 6

Moments

In statistics, raw moments quantify the shape of a dataset by emphasizing

deviations from the mean. The (r)-th raw moment, denoted as ( Mr ), is calculated
as:

r
Mr = Σ(Xi - X̄ ) / n

where:

● ( Mr ) is the (r)-th raw moment,

● ( n ) is the total number of data points,

● ( Xi ) is the (i)-th data point,

● ( X̄ ) is the mean of the dataset.

This report explores the calculation and significance of the first four raw
moments.

First-Order Raw Moment

The first-order raw moment ( r = 1 ) is given by:

M1 = Σ(Xi - X̄ ) / n
This moment measures the average deviation of each data point from the mean,
providing insights into the central tendency of the dataset.

Second-Order Raw Moment

The second-order raw moment ( r = 2 ) is defined as:

2
M2 = Σ(Xi - X̄ ) / n

This moment quantifies the variability or spread of the dataset, emphasizing


squared deviations from the mean.

Third-Order Raw Moment

The third-order raw moment ( r = 3 ) is calculated by:

3
M3 = Σ(Xi - X̄ ) / n

This moment captures the skewness of the dataset, indicating whether the
distribution is symmetric or skewed.

Fourth-Order Raw Moment

The fourth-order raw moment ( r = 4 ) is expressed as:


4
M4 = Σ(Xi - X̄ ) / n

This moment provides information about the kurtosis, highlighting the tails and
peakedness of the distribution.

Example: Exam Scores Dataset

Consider a larger dataset of exam scores: {60, 75, 80, 85, 90, 95, 100, 105, 110,
115}. Let's calculate the first four raw moments for this dataset using the provided
formula.

1. First-Order Raw Moment:

M1 = Σ(60 - X̄ ) + (75 - X̄ ) + ... + (115 - X̄ ) / 10

1. Second-Order Raw Moment:

2 2 2
M2 = Σ(60 - X̄ ) + (75 - X̄ ) + ... + (115 - X̄ ) / 10

1. Third-Order Raw Moment:

3 3 3
M3 = Σ(60 - X̄ ) + (75 - X̄ ) + ... + (115 - X̄ ) / 10

1. Fourth-Order Raw Moment:

4 4 4
M4 = Σ(60 - X̄ ) + (75 - X̄ ) + ... + (115 - X̄ ) / 10
r-th Moment about the Origin (O)

The r-th moment about the origin is given by:

r
Mr(O) = Σ(Xi ) / n

These moments describe the distribution of data points with respect to the origin,
providing insights into symmetry and concentration.

r-th Moment about Arbitrary Origin (A)

The r-th moment about an arbitrary origin A is expressed as:

r
Mr^(A) = Σ(Xi - A) / n

These moments offer insights into the distribution of data points relative to the
chosen origin A, allowing for a more flexible analysis.

Dimensionless Forms: Skewness and Kurtosis

The skewness (Sk) and kurtosis (K) can be expressed in dimensionless form:

(3/2)
SK = M3 / (M2)

2
K = M4 / (M2) - 3
These dimensionless measures provide standardized indicators of skewness and
kurtosis, making them comparable across different datasets.

Conditions for Skewness and Kurtosis

1. Skewness (SK):

2. If SK = 0, the distribution is perfectly symmetrical.

3. If SK > 0, the distribution is positively skewed (tail on the right).

4. If SK < 0, the distribution is negatively skewed (tail on the left).

5. Kurtosis (K):

6. If K = 0, the distribution has the same kurtosis as a normal distribution

(mesokurtic).
7. If K > 0, the distribution is leptokurtic (heavier tails and a sharper peak).

8. If K < 0, the distribution is platykurtic (lighter tails and a flatter peak).

Example:

Dataset of exam scores: {60, 75, 80, 85, 90, 95, 100, 105, 110, 115}. Let's
calculate the first four raw moments, r-th moment about the origin (O), r-th
moment about an arbitrary origin (A = 90), and dimensionless forms of skewness
and kurtosis.

1. First-Order Raw Moment: M1 = Σ(Xi - X̄ ) / 10

2
2. Second-Order Raw Moment: M2 = Σ(Xi - X̄ ) / 10
3
3. Third-Order Raw Moment: M3 = Σ(Xi - X̄ ) / 10

4
4. Fourth-Order Raw Moment: M4 = Σ(Xi - X̄ ) / 10

r
5. r-th Moment about the Origin (O): Mr(O) = Σ(Xi ) / 10

r
6. r-th Moment about Arbitrary Origin (A = 90): Mr(90) = Σ(Xi - 90) / 10

(3/2)
7. Dimensionless Skewness (SK): SK = M3 / (M2)

2
8. Dimensionless Kurtosis (K): K = M4 / (M2) - 3

You might also like