This volume, second in the series, provides essential background material for policy analysts, researchers, statisticians, and others interested in the application of microsimulation techniques to develop estimates of the costs and population impacts of proposed changes in government policies ranging from welfare to retirement income to health care to taxes. The material spans data inputs to models, design and computer implementation of models, validation of model outputs, and model documentation. Table of Contents Front Matter Introduction Model Design Model Evaluation Dynamic Simulation of Income Model 2 (DYNASIM2) Social Policy Simulation Database/Model (SPSD/M) REFERENCES Databases and Methods of Data Enhancement INTRODUCTION DATA NEEDS FOR MODELING INCOME SUPPORT PROGRAMS SURVEY DESIGN OF THE MARCH CPS AND THE SIPP Population Coverage Household and Individual Nonresponse Item Nonresponse Reporting Errors Sample Size Data Delivery Income Accounting Period Income Detail Households and Families Versus Program Filing Units Asset Holdings Other Variables for Related Policy Issues EVALUATION OF ESTIMATES OF PROGRAM PARTICIPANTS FROM CPS AND SIPP THE DIVISION OF LABOR FOR PRODUCTION OF DATABASES FROM CPS AND SIPP THE SUITABILITY OF CPS AND SIPP FOR INTEGRATED MODELING OF TAXES AND TRANSFERS CONCLUSIONS REFERENCES INTRODUCTION File Treatment Constrained and Unconstrained Statistical Matching Choosing the Matching Variables The EM-AF Statistical Match Merge File of the Office of Tax Analysis 1966 Merge File for Household Income Data Statistics Canada SCF-FEX Match AFDC-Census Match Conditional Independence Limitations in Modeling Reweighting of File B Data Resulting From Statistical Matching Iterative Proportional Fitting More Data Collection Multiple Matching and File Concatenation Rough Sensitivity Analysis CONCLUDING NOTE REFERENCES Model Design INTRODUCTION CONVERTING ANNUAL TO MONTHLY VALUES IN TRIM2, MATH, AND HITSM HITSM TRIM2 MATH HITSM TRIM2 Input Data MATH Comment Evaluation of Alternative Months Procedures Need for Further Validation PARTICIPATION FUNCTIONS FOR SSI, AFDC, AND FOOD STAMPS IN TRIM2, MATH, AND HITSM TRIM2 HITSM TRIM2 TRIM2 MATH Evaluation of Alternative Participation Functions APPENDIX: PERFORMANCE OF THE AFDC CALIBRATION PROCESS IN TRIM2 REFERENCES THE MODELS STARTING DATABASES SIMULATION OF LONGITUDINAL HISTORIES Labor Force Simulations PRISM Pension Coverage and Characteristics Interactions Between Pension and Social Security Eligibility and Retirement in PRISM Labor Force Simulations Retirement in DYNASIM2 DYNASIM2 Pension Coverage and Characteristics CONCLUSIONS REFERENCES Computing Technology INTRODUCTION Layered Structure Single Period Versus Future Projections Static Versus Dynamic Simulation Historical Background History Database Creation Database Structure and Size SPSD/M Parameters Model Execution Comparing Model Runs Output Facilities Other Software Facilities Availability, Customer Base, and Technical Support Future Work History Database Creation Database Structure and Size Database Adjustment Operating Characteristics TRIM2 Parameters Model Execution Output Facilities Other Software Facilities Operating Environment Comparison Background Informed Industry Observers Specific Technical Product Information Advances in Computer Processors Advances in Primary Memory Advances in Computer Systems Architecture Economic Studies of Industry Performance Assessment and Predictions Nature of the Demand for Microanalytic Simulation Models Factors Affecting the Availability of Microsimulation Models Functions of a System for Microanalytic Simulation Larger Model Execution on the Desktop Advances in Software Methodology Model Specification Object-Oriented Computing Simulation Module Consistency Advantages of Desktop-Based Environments Summary RECOMMENDATIONS AND CONCLUSIONS Short-Term Issues Porting TRIM2 to an MS-DOS Environment Translating TRIM2 Modules to the C Language Benefits of Porting TRIM2 Medium-Term Issues REFERENCES AND BIBLIOGRAPHY Model Validation 6 Variance Estimation of Microsimulation Models Through Sample Reuse COMPUTATIONS OF MICROSIMULATION MODELS NONPARAMETRIC VARIANCE ESTIMATION MODELS AND VARIABILITY BOOTSTRAPPING MICROSIMULATION MODELS Incorporation of Sensitivity Analyses - Measurement of Total Uncertainty Number of Replications Needed Use of Resampling Techniques in Constructing Confidence Intervals REFERENCES 7 Evaluations of Microsimulation Models: Literature Review HENDRICKS AND HOLDEN (1976A) HENDRICKS AND HOLDEN (1976B) GENERAL ACCOUNTING OFFICE (1977) HOLDEN (1977) HAYES (1982) JEFFERSON (1983) HAVEMAN AND LACKER (1984) ICF, INC. (1987) KORMENDI AND MEGUIRE (1988) BETSON (1988) DOYLE AND TRIPPE (1989) BEEBOUT AND HAWORTH (1989) BURTLESS (1989) DISCUSSION REFERENCES 8 A Validation Experiment with TRIM2 OVERVIEW OF THE EXPERIMENT CHOICE OF MODEL YEAR, PROGRAM YEAR, AND COMPARISON VALUES CHOICE OF COMPONENT MODULES AND ALTERNATIVES FOR STUDY IN TRIM2 OUTPUTS EXAMINED Total Variability Due to Use of Alternative Modules Analysis of Variance Methods for Sensitivity Analysis and External Validation Nonparametric Analysis Analysis of Categorical Data MAJOR CONCLUSIONS LIMITATIONS OF THE PRESENT STUDY DESIGN OF EXPERIMENT ISSUES REFERENCES INTRODUCTION NATURE OF POPULATION PROJECTION MODELS VALIDATING POPULATION PROJECTIONS Sensitivity Analysis External Validation CONFIDENCE INTERVALS CONCLUSIONS REFERENCES Model Documentation INTRODUCTION PURPOSES OF DOCUMENTATION AND EVALUATION CRITERIA Critique Suggestions Critique Suggestions Critique Suggestions COMPARISONS WITH IEEE STANDARDS REFERENCES

Table of Contents:
1 Front Matter; 2 Contents; 3 Introduction; 4 Databases and Methods of Data Enhancement; 5 Model Design; 6 Computing Technology; 7 Model Validation; 8 Model Documentation