双向固定效应和双重差分.ppt

Need larger set that makes sense Two options (many more) Cluster on state Cluster on state/disability * . gen disabled_state=100*disabled+statefip; reg wkswork1 _Ia* _Iy* _Ir* white black hispanic lths hsgrad somecol disabled treatment, cluster(statefip); .reg wkswork1 _Ia* _Iy* _Ir* white black hispanic lths hsgrad somecol disabled treatment, cluster(disabled_state); * Summary of results for cluster Coefficient on treatment (standard error) Regular OLS -1.998 (0.315) Cluster by state: -1.998 (0.487) Cluster by state/disab. -1.998 (0.532) * Linden and Rockoff Megan Kanka 7 year old girls Raped and murdered by neighbor who was convicted sex offender in 1981 Attack on 5 year old Attempted assault on s7-year old Lead to passage of “Megan’s Law” * Megan’s Law Sexual Offender (Jacob Wetterling) Act of 1994 Sexual offenders required to notify state of change of address Time limits vary across states (10 years after conviction or life) Required of all child sex offenders, some states require of all offenders * Megan’s Law 1996 Amendment to original law required states to publicly announce location and type of offense of sex offenders Indiana site /indiana.php * Economic question Crime negatively impacts property values Problem: crime is not random and neither are home purchases Therefore, getting an estimate of the impact of crime on housing prices is tough Megan’s law Sex offenders will most likely live in poorer areas How to separate the impact from their * Methodology Compare house sales in neighborhoods before and after arrival of sex offender Impact should be “local” so comparison sample is those on the same neighborhood but not near the offender * NC registry Between 1/1/1996 – 3/9/2003 A total of 8287 released offenders required to register 1007 left the state Of the remaining, 103 (1.4 percent) failed to register * Data Location/timing of sex offenders address Matched to home sales data – Charlotte/Mecklenburg county 1994-2004 Detailed characteristics