Accounting & Finance Training | Regression Analysis Training in Albany, NY

Regression Analysis Training Course Outline

Module 1: Introduction

• What is Regression Analysis?
• Publicly Available Data Sets
• Selected Application of Regression Analysis
• Step in Regression Analysis

Module 2: Simple Linear Regression

• Introduction
• Covariance and Correlation Coefficient
• Simple Linear Regression Model
• Parameter Estimation
• Test of Hypotheses
• Confidence Intervals
• Predictions
• Measuring the Quality of Fit
• Regression Line Through the Origin
• Trivial Regression Models

Module 3: Multiple Linear Regression

• Introduction
• Description of the Data and Model
• Parameter Estimation
• Interpretations of Regression Coefficients
• Centring and Scaling
• Properties of the Least Squares Estimators
• Multiple Correlation Coefficient
• Inference for Individual Regression Coefficient
• Test of Hypotheses in a Linear Model
• Predictions

Module 4: Regression Diagnostics: Detection of Model Violations

• Standard Regression Assumptions
• Various Types of Residuals
• Graphical Methods
• Graphs Before Fitting a Model
• Graphs After Fitting a Model
• Checking Linearity and Normality Assumptions
• Leverage, Influence, and Outliers
• Measures of Influence
• Potential-Residual Plot
• What to do with the Outliers?
• Role of Variables in a Regression Equation
• Effects of an Additional Predictor

Module 5: Qualitative Variables as Predictors

• Introduction
• Salary Survey Data
• Interaction Variables
• System of Regression Equations
• Other Application of Indicator Variables
• Seasonality

Module 6: Transformation of Variables

• Introduction
• Transformations to Achieve Linearity
• Bacteria Deaths Due to X-Ray Radiation
• Transformations to Stabilise Variance
• Detection of Heteroscedastic Errors
• Removal of Heteroscedasticity
• Weighted Least Squares
• Logarithmic Transformation of Data
• Power Transformation

Module 7: Problem of Correlated Errors

• Introduction: Autocorrelation
• Consumer Expenditure and Money Stock
• Durbin-Watson Statistic
• Removal of Autocorrelation by Transform
• Iterative Estimation with Autocorrelated Errors
• Autocorrelation and Missing Variables
• Analysis of Housing Starts
• Limitations of the Durbin-Watson Statistic
• Indicator Variables to Remove Seasonal

Module 8: Working with Collinear Data

• Principal Components
• Computations Using Principal Components
• Imposing Constraints
• Searching for Linear Functions
• Biased Estimation of Regression Coefficients
• Principal Components Regression
• Reduction of Collinearity in the Estimation Data
• Constraints on the Regression Coefficients
• Principal Components Regression: A Caution
• Ridge Regression
• Estimation by the Ridge Method
• Ridge Regression: Some Remarks