Introduction to Statistics - Download as PowerPoint by HC12100311334


									B AD 6243: Applied Univariate Statistics

     Introduction to Statistical

         Professor Laku Chidambaram
           Price College of Business
            University of Oklahoma
 Quantitative Research Questions
• Descriptive:
  – What is the GPA of the incoming MBA class?
• Difference:
  – Does the salary of new grads differ based on major?
• Relationship:
  – What is the relationship between gender and income
    in professional sports?
• Prediction:
  – Can we predict how much a manager will earn based
    on his education and level of experience?

                  BAD 6243: Applied Univariate Statistics   2
        Measurement Issues
• What is measurement?
  – Mapping from one set to another
  – Involves instruments and attributes
• Attributes of good measures
  – Reliability
  – Validity
  – Comparability
  – Objectivity
  – Quantifiability
                 BAD 6243: Applied Univariate Statistics   3
             Types of Variables
• Categorical/Nominal
   – Values have no intrinsic meaning (e.g., 1=Democrat;
• Ordinal
   – Rank ordering (e.g., “seed” in Wimbledon)
• Interval
   – Equal distance in scales means equal distance in
     property being measured (e.g., years, grades)
• Ratio
   – Values represent equal amounts from an absolute
     zero (e.g., height, weight etc.)
                   BAD 6243: Applied Univariate Statistics   4
               Inferential Statistics

                  Population                                      Sample

                  Population parameters                     Sample statistics
                             μ                                   x
                             σ                                   s
                             σ2                                  s2
                             ρ                                   r

• More random the sample, more generalizable the results
• Larger the sample, more precise the estimate of the population parameter
                       BAD 6243: Applied Univariate Statistics                  5
            Modeling Issues
• What is a model?
• Fit: Statistical model represents observed
  data well
• Abundance of linear models
• Models: Relations among variables
  – Magnitude of effect (strength -> part of overall
    available differentiation that is common)
  – Significance of effect (sample size)
                BAD 6243: Applied Univariate Statistics   6

To top