Longitudinal Analysis (Statistical Associates Blue Book Series 39)

Longitudinal Analysis (Statistical Associates Blue Book Series 39)

An introductory graduate level text on longitudinal analysis using SPSS, SAS, and Stata.<br /><br />328 pages<br /><br />Longitudinal analysis is an umbrella term for a variety of statistical procedures which deal with any type of data which is measured over time. Sections of this volume group longitudinal analysis methods under the following categories:<br /><br />Time series analysis, often used for projecting economic or other time series, with or without additional independent variables. Includes ARIMA models.<br /><br />Linear regression models, which incorporate time as an independent variable.<br /><br />Panel data regression models, <br /><br />Repeated measures GLM, used to implement analysis of variance and regression models.<br /><br />General estimating equations analysis (GEE), used to implement nonlinear forms of regression modeling, including logistic and probit regression for repeated measures data.<br /><br />Linear mixed modeling (LMM), used for multilevel analysis where multiple time periods are treated as a data level.<br /><br />Generalized linear mixed models for longitudinal data (GLMM), used to implement nonlinear forms of linear mixed modeling<br /><br />Structural equation modeling (SEM), used for growth curve analysis and modeling change in structural relationships across a limited number of time periods.<br /><br />Overview13<br />Comparing time series procedures13<br />GLM (OLS regression or ANOVA) with time as a variable13<br />Time series analysis (ex., ARIMA14<br />Repeated measures GLM14<br />Generalized estimating equations (GEE)14<br />Population-averaged panel data regression14<br />Random effects panel data regression15<br />Linear mixed models (LMM)15<br />Generalized linear mixed models (GLMM)15<br />Structural equation modeling15<br />GLMM-SEM15<br />Key concepts and terms16<br />Types of time-related data16<br />Statistical procedures for different types of data collected over time18<br />Time series analysis19<br />Overview19<br />Key Terms and Concepts19<br />Simple time series design20<br />Time series effects20<br />Serial dependence20<br />Stationarity20<br />Differencing21<br />Specification21<br />Autocorrelation21<br />Decomposition22<br />Model order22<br />Exponential Smoothing23<br />Overview23<br />Weighting23<br />Example24<br />Sequence charts24<br />Requesting exponential smoothing in SPSS26<br />Exponential smoothing model types: Simple27<br />Exponential smoothing model types: Holt's linear trend30<br />Exponential smoothing model types: Brown's linear trend31<br />Exponential smoothing model types: Damped trend32<br />Exponential smoothing model types: Seasonal effects32<br />Transformation of the dependent variable33<br />Statistical output for time series analysis in SPSS33<br />Residual and partial residual autocorrelation36<br />Displaying forecast values37<br />Saving exponential smoothing values in SPSS38<br />ARIMA Models40<br />Overview40<br />Example40<br />Constants and predictors41<br />Stationarity41<br />ARIMA p, d, and q parameters46<br />Types of ARIMA models50<br />Unit roots52<br />ARIMA for the example data52<br />Forecasts54<br />Residual Analysis55<br />Seasonal ARIMA61<br />ARIMA Modeling: Intervention and transfer function analysis62<br />The SPSS "Expert Modeler"68<br />Overview68<br />The “Expert Modeler” interface68<br />Leading indicator (CCF) analysis71<br />Overview71<br />SPSS set-up71<br />CCF output72<br />Creating a leading indicator variable74<br />Assumptions of time series analysis75<br />Stationarity75<br />Normally distributed independent residuals with homogenous variance76<br />Inconsequential outliers76<br />Frequently asked questions about time series analysis76<br />How many time periods are needed?76<br />What should the researcher do about missing data?76<br />When I try to specify p, d, and q for an ARIMA model, should non-significant spikes be treated as zero?77<br />I suspect there is not a single trend line but rather the trend is different for different subgroups in my population. How do I handle this?77<br />How does one go about disentangling age, period, and cohort time series effects?79<br />Is there an acceptable ARIMA model for all data?79<br />What is an ARFIMA model?80<br />Regression time series models80<br />Curve fitting80<br />Curve Estimation dialog in SPSS80<br />and 248 more pages of topics.