Statistical Methods for Data Reduction and Classification

Linear Regression

Linear Regression: Finds the best line that summarizes the relationship between two variables. (Imagine a bunch of dots and a line representing the relationship).

Dimensionality Reduction

A large number of variables results in a dispersion matrix that is too large to study. Dimensionality reduction reduces the number of variables to a few. Why?

  • Simpler Analysis: Fewer features make it easier to find patterns.
  • Faster Processing (e.g., Trading).
  • Better Visualization: It’s easier to visualize
Read More

Essential Geometric Definitions and Theorems

Angles and Lines

  • Acute Angle: An angle that measures less than 90º.
  • Angles: Formed by two rays that share a common endpoint, provided that the two rays are non-collinear.
  • Angle Bisector: A ray that contains the vertex and divides the angle into two congruent angles.
  • Complementary Angle: A pair of angles that sum 90º.
  • Congruent Angles: Two angles are congruent if and only if they have equal measure.
  • Corresponding Angles: Angles that are created at the same location at each intersection where a transversal
Read More

Understanding Vector Spaces, Linear Algebra, Eigenvalues, and Quadratic Forms

A.1.7. Vector Subspaces

A subset H of a vector space V is a vector subspace if and only if:

  • The zero vector 0 is in H.
  • For any vectors v1 and v2 in H, their sum v1 + v2 is also in H.
  • For any vector v1 in H and any scalar k, the scalar multiple k*v1 is also in H.

Examples of non-vector subspaces:

  • Sets defined by polynomial equations.
  • Vectors u = (x, y) where xy + x = 0.
  • Logarithmic functions.

Examples of vector subspaces:

  • Sets defined by linear equations like Ax + By + Cz = 0.
  • Sets defined by linear equations
Read More

Understanding Floating-Point Representation and Errors

Floating-Point Representation and Errors

t: precision – a positive integer

β: base (or radix) – an integer ≥ 2 (2, 10, 16)

e: exponent – an integer

(decimal) value d1.d2d3 · · · dt × β -> (d1 + d2/β1 + · · · + dt/βt-1 ) × βe

exponent range emin ≤ e ≤ emax

1 + 2*((B -1)B(t-1) * (emax- emin + 1)) norm

1 + 2 * (Bt * (emax – emin + 1)) denorm

Memory stored in 3 fields: sign (1 bit positive negative), exponent (depends on range), fraction or significant (depends on precision)

1 + EpsM

Read More

Key Marketing Research Variables and Editing

Scales of Measurement in Marketing Research

Understanding different scales is crucial for accurate data interpretation.

  • Nominal Scale: Numbers serve only as labels or tags for identifying and classifying objects. In marketing research, it’s used to identify respondents, brands, attributes, and other objects. Example: Numbers assigned to runners in a race.
  • Ordinal Scale: Indicates rank order, providing directional information in addition to nominal information. It measures non-numeric concepts like
Read More

Statistics and Sampling: True or False Practice Questions

1) A population is a set that includes all elements about which we wish to draw a conclusion. True

2) If we examine some of the population measurements, we are conducting a census of the population. False. Example: A census is defined as examining all of the population measurements.

3) A random sample is selected so that every element in the population has the same chance of being included in the sample. True

4) An example of a quantitative variable is the manufacturer of a car. False

5) An example

Read More