Fundamentals of Statistics contains material of various lectures and courses of H. Lohninger on statistics, data analysis and here for more.

Data Matrices

When you are measuring data in your lab, you finally have a data matrix which contains a certain number of rows and columns. The data is usually arranged in such a way, that each row of the data matrix contains all measurements on a single data object. The rows therefore represent the objects, or samples, whereas the columns represent the variables, or descriptors.

Example: Suppose you are an analytical chemist who is doing research on volcanic rocks sampled from Mt. Vesuvius near Naples, Italy. The rock samples collected from 43 sites at and around Mt. Vesuvius are analyzed for their iron, manganese, vanadium and fluorine content. The results are compiled in a data matrix with 43 rows and 4 columns:

The use of matrices for the representation of data has several advantages over other approaches, among which is the broad spectrum of readily available data analysis algorithms which can be applied directly to the data matrices. These algorithms are based on well developed matrix algebra.