Conley, Timothy G. and Giorgio Topa, "Socio-Economic Distance and Spatial Patterns in Unemployment", Journal of Applied Econometrics, Vol. 17, No. 4, 2002, pp. 303-327. There are two data files and 28 Matlab program files. They are all ASCII files in DOS format. The data files are zipped in the file ct-data.zip, and the program files are zipped in the file ct-programs.zip. Unix users whould use "unzip -a". DATA FILES chi8090.dat is a tab delimited ascii file, with 841 records. Each record has 103 entries (there are 103 variables). The first row in the data file contains the names of the variables. Each observation refers to a Census tract in the city of Chicago. The data come from file STF3a of the decennial censuses, 1980 and 1990 (summary information at the Census tract level) For each variable "var_name", three versions may exist: var_name 1980 Census data var_name9 1990 Census data var_nameD 1990-80 Census data long longitude of the centroid of the tract lat latitude of the centroid of the tract tract tract code from the Census Bureau pers100 100% count of persons in the tract per16up Number of persons 16 years and over in the tract pcamicd Percent Native Americans in the tract pcaspcd Percent Asian and Pacific Islanders in the tract pcblacd Percent Blacks in the tract pcspacd Percent Hispanics in the tract pcwhicd Percent Whites in the tract pcgerm Percent of German origin in the tract pcirish Percent of Irish origin in the tract pcital Percent of Italian origin in the tract pcpole Percent of Polish origin in the tract pcexecu Percent employed persons 16 yrs. and over in executive, administrative, and managerial occupations pcprofe Percent employed persons 16 yrs. and over in professional specialty occupations pctechn Percent employed persons 16 yrs. and over in technical and related support occupations pcsales Percent employed persons 16 yrs. and over in sales occupations pcadmin Percent employed persons 16 yrs. and over in administrative support occupations, including clerical pcprihs Percent employed persons 16 yrs. and over in private household occupations pcprote Percent employed persons 16 yrs. and over in protective service occupations pcservi Percent employed persons 16 yrs. and over in service occupations, except protective and household pcfarmi Percent employed persons 16 yrs. and over in farming, forestry, and fishing occupations pcpreci Percent employed persons 16 yrs. and over in precision production, craft and repair occupations pcmachi Percent employed persons 16 yrs. and over in machine operator, assembly, and inspection occupations pctrans Percent employed persons 16 yrs. and over in transportation and material moving occupations pchandl Percent employed persons 16 yrs. and over in handling, cleaning occupations unempr Percentage of unemployed persons over the civilian labor force (16 years and over). Unemployed persons are individuals who were neither "at work" nor "with a job but not at work" during the Census reference week, and who were actively looking for work during the last four weeks prior to the reference week segr segregation index: Euclidean distance of tract's racial composition from city-wide average pcnowhi Percent Non-Whites in the tract pchigh Percent persons 16 yrs. and over with at least a high school diploma pcclge Percent persons 16 yrs. and over with at least a college degree pperhh Average number of persons per household in the tract pcvac Percent of all housing units in the tract that are vacant pcmnger Percent employed persons 16 yrs. and over in managerial and professional specialty occupations mgroren median gross rent, in 1979 US$, of renter-occupied housing units vperhat Average housing value, in 1979 US$ (thousands), of owner-occupied non-condominium housing units pcolf_m Percent males 16 years and over who are out of the labor force pcolf_f Percent females 16 years and over who are out of the labor force pcfem Percent persons who are female pc018 Percent persons between 0 and 18 years of age pc024 Percent persons between 0 and 24 years of age pc1824 Percent persons between 18 and 24 years of age tmwkme Median travel time to work, in minutes, for workers 16 yrs. and over who do not work at home The file varnames.dat contains a list of the variable names in chi8090.dat. travel.dat contains data that were constructed directly by the authors, using published public transportation schedules from the CTA (Chicago Transit Authority). It is simply an 841 x 841 matrix of entries, where each entry (i,j) reports the travel time distance (in minutes) between Census tracts i and j in the city of Chicago. There are 841 records, and each record contains 841 entries: MATLAB PROGRAM FILES In Matlab, one should import the ascii data files described above and save then as .mat files. In particular, the user should name the matrix of travel time distances in travel.dat as a matrix D_time. Then, da_distance.m uses (long,lat) in chi8090.mat to make d_new: this is an (841x841) matrix of pairwise physical distances between Census tracts centroids The user should save d_time and d_new in a .mat file called tim13101.mat da_mkd_eth.m uses chi8090.mat to make d_eth81,91: these are racial/ethnic distance matrices: save them in a file timeth.mat also, same .m file uses chi8090.mat to make d_occ8,9: these are occupational distance matrices: save in timocc.mat % 2D figures: a1p_p89d.m uses tim13101.mat and chi8090.mat to make acfp.mat a1p_t89d.m uses tim13101.mat and chi8090.mat to make acft.mat a1p_e89d.m uses timeth.mat and chi8090.mat to make acfe.mat a1p_o89d.m uses timocc.mat and chi8090.mat to make acfo.mat mkrndr2.m uses tim13101.mat, timeth.mat, timocc.mat, chi8090.mat to make acfbdp.mat, acfbdt.mat, acfbde.mat, acfbdo.mat figur3.m uses acfp.mat and acfbdp.mat to make Fig.3 figurA1.m uses acft.mat and acfbdt.mat to make Fig.A1 figur4.m uses acfe.mat and acfbde.mat to make Fig.4 figurA2.m uses acfo.mat and acfbdo.mat to make Fig.A2 % 3D figures: acfs3d.m uses tim13101.mat, timeth.mat, timocc.mat, chi8090.mat to make acfupe8.mat, acfupe9.mat, acfucpe8.mat acfpe8.mat, acfpe9.mat, acfcpe8.mat acfuoe8.mat, acfuoe9.mat, acfucoe8.mat acfoe8.mat, acfoe9.mat, acfcoe8.mat acfupo8.mat, acfupo9.mat, acfucpo8.mat acfpo8.mat, acfpo9.mat, acfcpo8.mat mkrnupe.m uses tim13101.mat, timeth.mat, chi8090.mat to make a3d_upe8.mat, a3d_upe9.mat, a3ducpe8.mat a3d_pe8.mat, a3d_pe9.mat, a3d_cpe8.mat mkrnuoe.m uses timocc.mat, timeth.mat, chi8090.mat to make a3d_uoe8.mat, a3d_uoe9.mat, a3ducoe8.mat a3d_oe8.mat, a3d_oe9.mat, a3d_coe8.mat mkrnupo.m uses tim13101.mat, timocc.mat, chi8090.mat to make a3d_upo8.mat, a3d_upo9.mat, a3ducpo8.mat a3d_po8.mat, a3d_po9.mat, a3d_cpo8.mat figur5.m uses acfupe8.mat, a3d_upe8.mat to make Fig. 5 figur6.m uses acfupo8.mat, a3d_upo8.mat to make Fig. 6 figur7.m uses acfuoe8.mat, a3d_uoe8.mat to make Fig. 7 a1dec_race.m uses tim13101.mat, timeth.mat, timocc.mat, chi8090.mat to make a1decpr9.mat, a1dectr9.mat, a1decer9.mat, a1decor9.mat a1dec_educ.m uses tim13101.mat, timeth.mat, timocc.mat, chi8090.mat to make a1decpd9.mat, a1dectd9.mat, a1deced9.mat, a1decod9.mat a1dec_spmis.m uses tim13101.mat, timeth.mat, timocc.mat, chi8090.mat to make a1decps9.mat, a1dects9.mat, a1deces9.mat, a1decos9.mat figur8.m uses a1decpr9.mat, a1dectr9.mat, a1decer9.mat, a1decor9.mat to make Fig. 8 figurA3.m uses a1decpd9.mat, a1dectd9.mat, a1deced9.mat, a1decod9.mat to make Fig. A3 figur9.m uses a1decps9.mat, a1dects9.mat, a1deces9.mat, a1decos9.mat to make Fig. 9 table23.m uses chi8090.mat to generate Tables 2 and 3 kregn.m function that performs Gaussian kernel regression of "y" on "x", at points "x0" with bandwidth "h" kreg3d.m function that performs Gaussian kernel regression of "y" on "x" and "z", at points "x0" and "z0" with bandwidths "h1" and "h2"