BostonHomicide          package:strucchange          R Documentation

_Y_o_u_t_h _H_o_m_i_c_i_d_e_s _i_n _B_o_s_t_o_n

_D_e_s_c_r_i_p_t_i_o_n:

     Data about the number of youth homicides in Boston during the
     `Boston Gun Project'-a policing initiative aiming at lowering
     homicide victimization among young people in Boston.

_U_s_a_g_e:

     data(BostonHomicide)

_F_o_r_m_a_t:

     A data frame containing 2 monthly time series and a factor coding
     seasonality.

     _h_o_m_i_c_i_d_e_s time series. Number of youth homicides.

     _p_o_p_u_l_a_t_i_o_n time series. Boston population (aged 25-44), linearly
          interpolated from annual data.

     _p_o_p_u_l_a_t_i_o_n_B_M time series. Population of black males (aged 15-24),
          linearly interpolated from annual data.

     _a_h_o_m_i_c_i_d_e_s_2_5 time series. Number of adult homicides (aged 25 and
          older).

     _a_h_o_m_i_c_i_d_e_s_3_5 time series. Number of adult homicides (aged 35-44).

     _u_n_e_m_p_l_o_y time series. Teen unemployment rate (in percent).

     _s_e_a_s_o_n factor coding the month.

     _y_e_a_r factor coding the year.

_D_e_t_a_i_l_s:

     The `Boston Gun Project' is a policing initiative aiming at
     lowering youth homicides in Boston. The project began in early
     1995 and implemented the so-called `Operation Ceasefire'
     intervention which began in the late spring of 1996.

     More information is available at:

     <URL: http://www.ksg.harvard.edu/criminaljustice/research/bgp.htm>

_S_o_u_r_c_e:

     Piehl et al. (2004), Figure 1, Figure 3, and Table 1.

     From the table it is not clear how the data should be linearly
     interpolated. Here, it was chosen to use the given observations
     for July of the corresponding year and then use 'approx' with
     'rule = 2'.

_R_e_f_e_r_e_n_c_e_s:

     Piehl A.M., Cooper S.J., Braga A.A., Kennedy D.M. (2003), Testing
     for Structural Breaks in the Evaluation of Programs, _The Review
     of Economics and Statistics_, *85*(3), 550-558.

     Kennedy D.M., Piehl A.M., Braga A.A. (1996), Youth Violence in
     Boston: Gun Markets, Serious Youth Offenders, and a Use-Reduction
     Strategy, _Law and Contemporary Problems_, *59*, 147-183.

_E_x_a_m_p_l_e_s:

     data(BostonHomicide)
     attach(BostonHomicide)

     ## data from Table 1
     tapply(homicides, year, mean)
     populationBM[0:6*12 + 7]
     tapply(ahomicides25, year, mean)
     tapply(ahomicides35, year, mean)
     population[0:6*12 + 7]
     unemploy[0:6*12 + 7]

     ## model A
     ## via OLS
     fmA <- lm(homicides ~ populationBM + season)
     anova(fmA)
     ## as GLM
     fmA1 <- glm(homicides ~ populationBM + season, family = poisson)
     anova(fmA1, test = "Chisq")

     ## model B & C
     fmB <- lm(homicides ~ populationBM + season + ahomicides25)
     fmC <- lm(homicides ~ populationBM + season + ahomicides25 + unemploy)

     detach(BostonHomicide)

