# Class Notes

You can find copies of the class notes here. To view and print them you must have Adobe Reader, a free program loaded in your machine. Normaly, you can just click on the notes and Reader will start.

You can obtain the program from the University's Technology page by looking for software. Or you can downloaded it directly from Adobe. Most University machines have this program installed.

• Introduction to applied statistics, desktop computing, the internet
• MINITAB and SPSS (to a lesser extent), saving files, worksheets.
• Small sample test of means, t distribution, critical value and region.
• Confidence intervals for means and differences of means, MINITAB procedures
• One-way and two-way frequency distributions, cross-classifications, chi square test of independence.
• Cross-classifications, chi square test, remarks about significance testing, strength of association, odds ratio, log odds ratio.
• Odds ratios for I X J tables; comments about the chi square test, linear regression, scatterplot, slope, intercept.
• Basic regression model, least squares idea, regression coefficients, interpretation of coefficients
• Causal inference with regression analysis, examples, OLS
• Inference for regression, t tests, analysis of variance, F tests, confidence intervals, multiple R, coefficient of determination.
• Short set of notes that discuss the properties of the multiple regression coefficient or coefficient of determination; that is, R-squared.
• Example of hypothesis test on regression coefficients, confidence intervals, correlation coefficient, interpretation of coefficients.
• Standardized regression (beta weights, standardized regression coefficients), multiple regression model, partial regression coefficients
• Extended example of multiple regression, t and F tests, confidence intervals, standardized regression coefficients, model building.
• Regression with categorical independent variables, dummy coding, "effects," interaction.
• Multiple regression with categorical variables, interaction, examples of F tests.
• Regression assumptions, residuals, standardized residuals, partial regression plots, plots of residuals versus fitted values
• Examples
• Simultaneous confidence intervals for partial regression coefficients, multicolinearity, colinearity, variance inflation factor (VIF), causation, randomization, experimental design, internal validity, external validity.
• Time series, intervention analysis, dummy variables, time counters, simple intervention models
• Time series, autocorrelation, lagged variables, Durbin-Watson statistic, intervention.
• Time series processes, autocorrelation, moving average, ARIMA, logistic regression.
• Models for log odds (logits), log odds, probabilities, linear probability model, complete separation of data, parameter interpretation.
• Logistic regression, inference, correctly classified rate, The Bell Curve.
• Strategies for building multiple regression models.

Applied Statistics page