Data dimensionality reduction in anthropometrical investigations

Area of Science:

• Anthropometry
• Biostatistics
• Data Science

Background:

• Decision-making and diagnosis often rely on analyzing large, diverse datasets.
• Principal Component Analysis (PCA) is a statistical technique for data dimensionality reduction.

Purpose of the Study:

• To simplify decision-making processes through data dimensionality reduction using PCA.
• To evaluate the impact of PCA on analyzing pubertal growth patterns.

Main Methods:

• Applied PCA to anthropometrical data from 400 boys.
• Calculated and interpreted three principal components.
• Approximated component variability using fourth-order polynomials to study growth changes.

Main Results:

• Dimensionality reduction via PCA resulted in approximately 25% information loss.
• The three principal components retained 75% of the total information.
• This retained information is sufficient for making informed decisions.

Conclusions:

• The PCA approach is effective for analyzing complex anthropometrical data.
• The method supports data-driven decision-making in the context of growth studies.