Presentation on theme: "Introduction: Correlation and Regression The General Linear Model is a phrase used to indicate a class of statistical models which include simple linear."— Presentation transcript:
Introduction: Correlation and Regression The General Linear Model is a phrase used to indicate a class of statistical models which include simple linear regression analysis. Regression is the predominant statistical tool used in the social sciences due to its simplicity and versatility. Also called Linear Regression Analysis. We will examine regression first, and then see how correlation is one portion of regression analysis.
Simple Linear Regression: The Basic Mathematical Model Regression is based on the concept of the simple proportional relationship - also known as the straight line. We can express this idea mathematically! Theoretical aside: All theoretical statements of relationship imply a mathematical theoretical structure. Just because it isn’t explicitly stated doesn’t mean that the math isn’t implicit in the language itself!
Simple Linear Relationships Alternate Mathematical Notation for the straight line - don’t ask why! 10th Grade Geometry Statistics Literature Econometrics Literature
Alternate Mathematical Notation for the Line These are all equivalent. We simply have to live with this inconsistency. We won’t use the geometric tradition, and so you just need to remember that B 0 and a are both the same thing.
Linear Regression: the Linguistic Interpretation In general terms, the linear model states that the dependent variable is directly proportional to the value of the independent variable. Thus if we state that some variable Y increases in direct proportion to some increase in X, we are stating a specific mathematical model of behavior - the linear model.
The linear model is represented by a simple picture
The Mathematical Interpretation of the Regression Parameters a = the intercept the point where the line crosses the Y-axis. (the value of the dependent variable when all of the independent variables = 0) b = the slope the increase in the dependent variable per unit change in the independent variable (also known as the 'rise over the run')
The Error Term Such models do not predict behavior perfectly. So we must add a component to adjust or compensate for the errors in prediction. Having fully described the linear model, there are now several courses to spend on the error term.
The 'Goal' of Ordinary Least Squares Ordinary Least Squares (OLS) is a method of finding the linear model which minimizes the sum of the squared errors. Such a model provides the best explanation/prediction of the data.
Why Least Squared error? Why not simply minimum error? It is similar to the problem with the average deviation The error’s about the line sum to 0.0! Minimum absolute deviation (error) models now exist, but they are mathematically cumbersome. Try algebra with | Absolute Value | signs! We square the error to get rid of the negative signs, and take the square root to get back to the “root mean squared error.” Which we don’t use very much Some feel that big errors should be more influential than small errors.
Other models are possible... Best parabola...? (i.e. nonlinear or curvilinear relationships) Best maximum likelihood model... ? Best expert system...? Complex Systems…? Chaos models Catastrophe models others
The Notion of Linear Change The linear aspect means that the same amount of increase unemployment will have the same effect on crime at both low and high unemployment. A nonlinear change would mean that as unemployment increased, its impact upon the crime rate might increase at higher unemployment levels.
Minimizing the Sum of Squared Errors Who put the Least in OLS In mathematical jargon we seek to minimize the Unexplained Sum of Squares (USS), where:
T-Tests Since we wish to make probability statements about our model, we must do tests of inference. Fortunately,
Measures of Goodness of fit The Correlation coefficient r-squared The F test
The correlation coefficient A measure of how close the residuals are to the regression line It ranges between -1.0 and +1.0 It is closely related to the slope.
Goodness of fit The correlation coefficient A measure of how close the residuals are to the regression line It ranges between -1.0 and +1.0 r 2 (r-square) The r-square (or R-square, or r 2 ) is also called the coefficient of determination Ranges between 0.0 and 1.0 Expresses the % of Y explained by X
Tests of Inference t-tests for coefficients F-test for entire model Since we are interested in how well the model performs at reducing error, we need to develop a means of assessing that error reduction. Since the mean of the dependent variable represents a good benchmark for comparing predictions, we calculate the improvement in the prediction of Yi relative to the mean of Y (the best guess of Y with no other information).