* Truncated Poisson regression to model lottery winners;

data lottery;

label

          popul  = 'population in 1000s'

          logpop = 'log population'

          area   = 'in square miles'

          mill   = 'property tax rate'

          dens   = 'population density(popul/area)';



town $ 1-14  winners   popul    area    mill

--------------------------------------------

Ansonia            6    17.9     6.2    28.9 

Beacon Falls       3     5.3     9.8    25.0 

Branford          11    28.0    27.9    22.6 

Cheshire           6    26.2    33.0    27.1 

Clinton            2    12.8    17.2    27.9 

Derby              6    12.0     5.3    29.6 

East Haven         9    26.5    12.6    37.1 

Guilford           6    20.3    47.7    28.6 

Hamden             9    52.0    33.0    34.1 

Madison            5    16.0    36.3    22.3 

Milford           10    49.5    23.5    30.8 

N. Branford        2    13.1    26.8    26.9 

North Haven       12    21.6    21.0    23.4 

Old Saybrook       1     9.3    18.3    15.3 

Orange             9    12.5    17.6    23.8 

Oxford             3     9.1    33.0    29.0 

Seymour            1    14.5    14.7    40.5 

Shelton            7    36.0    31.4    21.6 

Trumbull          14    33.0    23.5    24.1 

West Haven        12    54.0    10.6    41.4 

Woodbridge         1     8.0    19.3    28.4