InGrid Summer School “Reaching out to hard-to-survey groups among the poor” HIVA-KU Leuven, Leuven - Belgium, 30 May -3 June 2016 Small Area Models for Linking Deprivation to Local Areas in Italy Gennaro PUNZO University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 1
IN ITALY, THE SOURCE OF OFFICIAL STATISTICS ON POVERTY, PROVIDED BY ISTAT, IS THE HOUS USEHOLD BUD UDGE GET SUR URVEY (HBS)... IT IS PLANNED TO COMBINE IT DOESN’T ALLOW TO PRODUCE TO ANALYSE THE THESE ASPECTS IN RELIABLE INCOM OME PEOPLE’S LIVING ORDER TO YIELD POVERTY ESTIMATES CONDITIONS ONLY AT A NATIONAL MORE RELIABLE ACCORDING TO A LEVEL OR, AT LEAST, MULTID IDIM IMENSIO IONAL POVERTY FOR LARGE AND FUZZY ZZY GEOGRAPHICAL ESTIMATES APPROACH DIVISIONS Poverty is considered as a latent con contin inuum uum (Lemmi et al., 1997) that is not directly observable and, in addition to the level of monetary income, it can be revealed by a variety of indicators of life-style deprivation University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 2
AIM OF OUR WORKS Exploring POVERTY PATTERNS and DIFFERENTIALS across Italian NUTS3 regions (administrative provinces) for several dimensions of life-style deprivation STEPS JOINT ANALYSIS OF “ MONETARY” AND “ SUPPLEMENTARY” DEPRIVATION ACCORDING TO A MULTIDIMENSIONAL AND FUZZY APPROACH MANIFEST AND LATENT DEPRIVATION MEASURES (BETTI ET AL., 2006; BETTI AND VERMA, 2006) BORROWING STRENGTH ACROSS SMALL AREAS, TIME, AND SPACE: FAY-HERRIOT MODEL (1979) EFFICIENCY RAO – YU MODEL (1992, 1994) GAIN PETRUCCI-SALVATI MODEL (2004) University of Naples “Parthenope” (Italy) 2nd June 2016 3 Department of Management and Quantitative Studies
THE MAIN PROBLEM OF POVERTY ESTIMATES AT A SUB - NATIONAL LEVEL IS THEIR HIGH LEVEL OF VARIABILITY due to the SAMPLING ERROR that increases with the decreasing size of su sub -samples in the areas (i.e., of the regions, or even at the level of smaller units, which in Italy are the PROVINCES) A NEW CLASS OF ESTIMATORS SMALL AREA ESTIMATION (SAE MODELS) University of Naples “Parthenope” (Italy) 2nd June 2016 4 Department of Management and Quantitative Studies
What’s a Small Area? As a rule, a domain is regarded as small ll if the domain-specific sample is not large enough to support direct estimates of adequate precision; they are likely to produce large standard errors due to the unduly small size of the sample in the area (Ghosh & Rao, 1994) therefore, in small areas… it is necessary to use special estimators that “borrow strength” from related areas across space and/or time or through auxiliary information that is supposed to be correlated to the variable of interest University of Naples “Parthenope” (Italy) 2nd June 2016 5 Department of Management and Quantitative Studies
SMALL AREA ESTIMATORS OVERESTIMATE THE DIRECT ESTIMATORS VARIABILITY AMONG (refer to the estimates derived SMALL AREAS from the survey data for (It is due to the effect of the sampling small areas concerned) error because of the smallness of the sample size available at small area level ) INDIRECT ESTIMATORS (based on models relating to UNDERESTIMATE THE the target variable to some TRUE VARIABILITY available auxiliary variables) University of Naples “Parthenope” (Italy) 2nd June 2016 6 Department of Management and Quantitative Studies
SMALL AREA ESTIMATORS COMPOSITE ESTIMATORS weighted mixture of DIRECT and SYNTHETIC estimators of the same unknown parameter 1) IT IS MORE LIKELY TO REFLECT THE TRUE VARIABILITY THAN EITHER OF THE TWO (DIRECT AND SYNTHETIC) 2) AS A RESULT, IT BALANCES THE POTENTIAL BIAS OF SYNTHETIC ESTIMATOR, WHICH IS CAPTURED BY THE MEAN-SQUARED ERROR, AGAINST THE HIGHER INSTABILITY OF THE DIRECT ONES University of Naples “Parthenope” (Italy) 2nd June 2016 7 Department of Management and Quantitative Studies
A FOCUS ON DEPRIVATION IN ITS MULTIPLE DIMENSIONS... FUZZY SET approach (Zadeh, 1965) and, in particular, TOTALLY FUZZY and RELATIVE method (Cheli-Lemmi, 1995) A NEW CLASS OF MONETARY AND SUPPLEMENTARY DEPRIVATION MEASURES TREATING POVERTY AS A MATTER OF DEGREE, REPLACING THE TRADITIONAL DICHOTOMIZATION POOR/NON POOR FUZZY MONETARY FUZZY SUPPLEMENTARY PROPENSITY TO PROPENSITY TO OVERALL INCOME POVERTY NON-MONETARY DEPRIVATION University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 8
SUPPLEMENTARY POVERTY MEASURES Following Betti–Verma (2004) and aggregating 24 basic non-monetary variables: OVERALL DEPRIVATION 1 FSU SUP Fuzzy Supplementary MEASURE (NON-MONETARY) Moreover, following Nolan–Whelan (1996), Whelan et al. (2001) and Betti–Verma (2004): FIVE DIMENSION–SPECIFIC DEPRIVATION MEASURES Basic life-style Lack of ability to afford most 2 FSUP_1 deprivation basic requirements (7 items) 1) Keeping the household’s principal Secondary life-style accommodation adequately warm 3 FSUP_2 deprivation 2) Paying for a week’s annual holiday away from home 3) Replacing any worn-out furniture 4) Buying new rather than second hand clothes 5) Eating meat chicken or fish every second day, “Enforced” absence of widely desired possession if the household wanted to because of lack of resources (6 items) 6) Having friends or family for a drink or meal at 1) Car or van least one a month 2) Colour TV 7) Inability to meet payment of scheduled 3) Video recorder mortgage payments, utility bills or hire 4) Micro wave purchase instalments 5) Dishwasher 6) Telephone University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 9
SUPPLEMENTARY POVERTY MEASURES FIVE DIMENSION–SPECIFIC DEPRIVATION MEASURES Lack of basic housing facilities (3 items) Housing 1) Bath or shower 4 FSUP_3 2) Indoor flushing toilet Facilities 3) Hot running water Housing Serious problems with accommodation (3 items) 5 FSUP_4 Deterioration 1) Leaky roof 2) Damp walls, floors, foundation, etc. Environmental 6 FSUP_5 3) Rot in window frames or floors problems Problems with the neighborhood and environment (5 items) 1) Shortage of space 2) Noise from neighbours or outside 3) Too dark/not enough light 4) Pollution, grime or other environmental problems caused by traffic or industry 5) Vandalism or crime in the area University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 10
SAE MODELS: FAY-HERRIOT MODEL In the sphere of SAE models, we choose… FIRST Area Level Random Effect Model (Fay-Herriot, 1979)... STEP considering the area random effects as INDEPENDENT UNDER THIS MODEL, THE EBLUP ESTIMATOR IS OBTAINED Punzo et al. (2007; 2011) EBLUP composite estimates of poverty measures at a provincial level (NUTS3) with advanced degrees of efficiency in comparison with the corresponding direct estimates University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 13
METHODOLOGY: SECOND STEP In order to assess the gain, in terms of efficiency, that could be achieved by borrowing strength across BOTH SMALL AREAS AND TIME... RAO AND YU MODEL (1992, 1994) as extension of the basic Fay–Herriot (1979) Survey data Auxiliary variables ECHP sample Istat Territorial (waves 1994 – 2001) Indicators DATA SOURCES DIRECT SYNTHETIC ESTIMATES ESTIMATES University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 14
AREA-LEVEL PROVINCIAL INDICATORS AS INDEPENDENT VARIABLES Activity R Rate 11 Crude D Death Rate 1 2 Em Employment R Rate 12 Infant M Mortality R Rate 3 Unemployment R Rate 13 Marriage R Rate 4 Popu pulation D n Dens nsity 14 Crime R Rate AUXILIARY Resident nt P Popu pulation pe n per 100 15 Suic icid ides p s per 1 100.000 in inhabit itants s VARIABLES 5 inha nhabi bitant nts MATRIX Index of of T Territor orial C Con oncentration on 6 16 Legal al S Separ arat ation Rat ate of the he R Resident nt P Popu pulation n 7 Net M Migratory R Rate 17 Divorce R Rates 8 Hosp spit italiz izatio ion R Rate 18 Gross D ss Domest stic ic Product ( (GDP) Growth En Enterprises Rate 9 Public ic H Hosp spit italiz izatio ion Rate 19 (net o of a agriculture) 10 Crude B Birth Rate STEPWISE PROCEDURE... University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 15
SAE MODELS: PETRUCCI-SALVATI In order to try to explain the portion of the random error unaccounted for and left over by exogenous variables (territorial indicators)... Spatial Area Level Random Effect Model THIRD (Petrucci–Salvati, 2004)... considering the area specific STEP random effects SPATIALLY CORRELATED UNDER THIS MODEL, THE SPATIAL EBLUP ESTIMATOR IS OBTAINED SEBLUP composite estimates of poverty measures at NUTS3 level with different degrees of efficiency in comparison with the corresponding direct and EBLUP estimates University of Naples “Parthenope” (Italy) 2nd June 2016 Department of Management and Quantitative Studies 17
18
Recommend
More recommend
