Title Spline regression models / Lawrence C. Marsh, David R. Cormier

Published Thousand Oaks, Calif. : Sage Publications, 2001

Online access available from: Sage Research Methods Online Books    View Resource Record

Copies

Description 1 online resource (vi, 69 pages) : illustrations Series Quantitative applications in the social sciences ; 07-137 Quantitative applications in the social sciences ; no. 07-137 Contents Polynomial Regression Models -- Spline Knot Locations Known in Advance -- Splines With Unknown Knot Locations -- Unknown Number of Spline Knots -- Introduction to Spline Models -- Interrupted Regression Analysis -- Piecewise Linear Regression -- Cubic Polynomial Regression -- Important Features of Spline Models -- Splines With Known Knot Locations -- Linear Spline Regression Models -- Quadratic and Higher Order Spline Regression Models -- Hybrid Spline Regression Models -- Model Comparison Issues -- Model Selection Criteria -- Polynomial Regression and Perfect Multicollinearity -- F Statistics and t Statistics -- Autocorrelation and the Durbin-Watson Statistic -- Splines With Unknown Knot Locations -- Transforming Discrete Response Into Continuous Measure -- Interrupted Regression Analysis -- Adjusting Intercepts Only -- Adjusting Intercepts and Slopes -- Splines With Known Knot Locations -- Unknown Spline Knot Location Estimation -- Quadratic Spline With Unknown Knot Locations -- The Wald Test -- Model Selection Conclusion -- Splines With an Unknown Number of Knots -- Stepwise Regression as a Powerful Nonparametric Method -- Determining the Number, Location, and Degree of the Spline Knots -- Smooth Splines for Long-Term Investing -- Moderately Sensitive Splines for Medium-Term Investing -- Highly Sensitive Splines for Short-Term Investing -- Spline Regression Forecasting -- SAS Program to Calculate Standard Error Summary Spline Regression Models shows how to use dummy variables to formulate and estimate spline regression models both in situations where the number and location of the spline knots are known in advance, and where estimation is required Bibliography Includes bibliographical references (pages 67-68) Notes Print version record Subject Regression analysis. Social sciences -- Statistical methods. Form Electronic book Author Cormier, David R. ISBN 1412985900 (ebook) 145221185X 9781412985901 (ebook) 9781452211855

  Permalink