Correlation and Regression Analysis Analysis. . Angie Daniela

Correlation and Regression Analysis Analysis. . Angie Daniela

1. Frequently used by market researchers

   • It studies the relationship between two or more variables.
   • According to Purpose
     • Correlation Analysis
       • Measures the closeness of the relationship between two or more variables on an interval interval scale.
         • The joint variation of the two measurements is not subject to restrictions by the experimenter.
       • Measurements in the Correlation Correlation Analysis.
         • Dependent variable (criterion)
         • Independent variable (predictor)
     • Regression Analysis
       • Used to derive an equation relating the criterion variable to one or more predictor variables.
       • Regression Analysis Yes
         • Only one predictor variable is used.
         • Poder explicativo de la regresión
           • Variation measures
             • Sum of Squares
               • Regression. -Error. -Total.
                 • They allow to calculate:
                   • Coefficient of determination.
                   • Correlation correlation.
                   • Standard error of the estimate
                   • ANOVA in regression
         • Assumptions
           • Residual Analysis
             • Allows examination of linearity and violation of assumptions.
           • -Y values are normally distributed for each X value.
           • -The random error is normal distributed with zero mean and constant variance.
           • -Homoscedasticity (constant variance). -Independence of the errors.
         • Population regression line
           • Parameters.
           • Recta de regresión muestral.
             • Estimaciones a través de EMV o Mínimos cuadrados.
       • Multiple Regression Analysis.
         • Two or more variables more predictor variables are used. Prediction.
           • ANOVA (Analysis of Variables)
             • Compare two or more measurements.
             • One-way (between-subjects factor)
             • Two-way (various factors between subjects)
             • For repeated measures (within-subjects factor)
             • Mixed (inter- and intra-subject factors)