PROTOTYPE DEVELOPMENT OF STATISTICAL LAND-USE PREDICTION MODEL FOR GREATER BOSTON REGION

TRAFFIC RESEARCH CORPORATION HAS DEVELOPED AND TESTED A PROTOTYPE MODEL FOR THE SUBREGIONAL DISTRIBUTION OF POPULATION AND EMPLOYMENT GROWTH IN A METROPOLITAN AREA. THE MODEL, CALLED THE EMPIRIC MODEL, REQUIRES EXTERNALLY SPECIFIED REGIONAL GROWTH TOTALS FOR POPULATION AND EMPLOYMENT CATEGORIES TO BE PROJECTED. THE DEVELOPMENT WORK WAS CARRIED OUT IN CONJUNCTION WITH THE EFFORTS OF THE BOSTON REGIONAL PLANNING PROJECT TO PREPARE A COMPREHENSIVE DEVELOPMENT PLAN FOR THE BOSTON REGION. VALIDITY CHECKS HAVE BEEN CARRIED OUT BY APPLYING THE MODEL TO FORECAST FIVE CATEGORIES OF POPULATION AND EMPLOYMENT FROM 1950 TO 1960 FOR 29 SUBREGIONS AND THEN COMPARING THE OBSERVED AND CALCULATED SUBREGIONAL ACTIVITY LEVELS FOR 1960. ROOT-MEAN- SQUARE ERROR RATIOS OBTAINED WITH PROTOTYPE EMPIRIC MODEL APPLICATION III WERE ABOUT 1 AND 3 PERCENT FOR TOTAL POPULATION AND TOTAL EMPLOYMENT, RESPECTIVELY. THE RESULTS SEEM TO CONFIRM THE HYPOTHESIS THAT URBAN LAND USE MAY BE PREDICTED ON A SUBREGIONAL LEVEL, USING AN ASSOCIATIVE OR STATISTICAL MODEL OF THE EMPIRIC TYPE, WITH SUFFICIENT ACCURACY FOR LAND-USE AND TRANSPORTATION PLANNING PURPOSES. THE USE OF SUCH A MODEL ENABLES PLANNERS TO STUDY SYSTEMATICALLY THE EFFECTS OF TRANSPORTATION FACILITIES, LAND-USE CONTROLS, AND OTHER POLICIES ON URBAN DEVELOPMENT AND TO PRODUCE STAGED PLANS AND POLICIES WHICH TAKE THESE INTERACTIONS INTO ACCOUNT. /AUTHOR/