Academia.eduAcademia.edu

Figure 39 – uploaded by Kyle Weber

See full PDFdownloadDownload figure
Table L3: The Impact of RTC Laws on the Murder Rate, DAW covariates, Full Sample, 1977-2014

Table 39 L3: The Impact of RTC Laws on the Murder Rate, DAW covariates, Full Sample, 1977-2014

Related Figures (92)

The decline in violent crime rates has been far greater in states with no RTC laws, 1977-2014 Note: Ilinois excluded since its concealed carry law did not go into effect until 2014. From 1977-2013, the violent crime rate in Illinois fell by 36 percent, from 631 to 403 crimes per 100,000 people.
The decline in violent crime rates has been far greater in states with no RTC laws, 1977-2014 Note: Ilinois excluded since its concealed carry law did not go into effect until 2014. From 1977-2013, the violent crime rate in Illinois fell by 36 percent, from 631 to 403 crimes per 100,000 people.
All models include year and state fixed effects, and OLS estimates are weighted by state population. Robust standard errors (clustered at the state level) are provided next to point estimates in parentheses. In Column 2 we present Incidence Rate Ratios (IRR) estimated using negative binomial regression, where population is included as a control variable, as STATA does not have a weighting function for nbreg. The null hypothesis is that the IRR equals 1. The crime data is from the Uniform Crime Reports (UCR). Six demographic variables (based on different age-sex-race categories) are included as controls in the regression above. Other controls include the lagged incarceration rate, the lagged police employee rate, real per capita personal income, the unemployment rate, poverty rate, beer, and percentage of the population living in MSAs. * p < .1, ** p < .05, *** p < .01. All figures reported in percentage terms. Table 3: Panel Data Estimates Suggesting that RTC Laws increase Violent and Property Crime: State and Year Fixed Effects, DAW Regressors, 1979-2014
All models include year and state fixed effects, and OLS estimates are weighted by state population. Robust standard errors (clustered at the state level) are provided next to point estimates in parentheses. In Column 2 we present Incidence Rate Ratios (IRR) estimated using negative binomial regression, where population is included as a control variable, as STATA does not have a weighting function for nbreg. The null hypothesis is that the IRR equals 1. The crime data is from the Uniform Crime Reports (UCR). Six demographic variables (based on different age-sex-race categories) are included as controls in the regression above. Other controls include the lagged incarceration rate, the lagged police employee rate, real per capita personal income, the unemployment rate, poverty rate, beer, and percentage of the population living in MSAs. * p < .1, ** p < .05, *** p < .01. All figures reported in percentage terms. Table 3: Panel Data Estimates Suggesting that RTC Laws increase Violent and Property Crime: State and Year Fixed Effects, DAW Regressors, 1979-2014
All models include year and state fixed effects, and OLS estimates are weighted by state population. Robust standard errors (clustered at the state level) are provided next to point estimates in parentheses. In Panel A, 36 demographic variables (based on different age-sex-race categories) are included as controls in the regressions above. In Panel B, only 6 demographic variables are included. In Panel C, only 6 demographic variables are included and controls are added for incarceration and police. For both Panels, other controls include the previous year’s violent or property crime arrest rate (depending on the crime category of the dependent variable), state population, population density, real per capita income, real per capita unemployment insurance payments, real per capita income maintenance payments, and real retirement payments per person over 65. * p < .1, ** p < .05, *** p < .O1. All figures reported in percentage terms.
All models include year and state fixed effects, and OLS estimates are weighted by state population. Robust standard errors (clustered at the state level) are provided next to point estimates in parentheses. In Panel A, 36 demographic variables (based on different age-sex-race categories) are included as controls in the regressions above. In Panel B, only 6 demographic variables are included. In Panel C, only 6 demographic variables are included and controls are added for incarceration and police. For both Panels, other controls include the previous year’s violent or property crime arrest rate (depending on the crime category of the dependent variable), state population, population density, real per capita income, real per capita unemployment insurance payments, real per capita income maintenance payments, and real retirement payments per person over 65. * p < .1, ** p < .05, *** p < .O1. All figures reported in percentage terms.
Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
erceived favorable experience with the law, we see why it can be quite easy to draw the inaccurat ausal inference that Texas’ crime decline was facilitated by its RTC law. The public may perceiv he fal ing crime rate post-1996 (the solid black line) but our analysis suggests that Texas woul lave experienced a more sizable violent crime decline if it had not passed an RTC law (the dotte ine). More specifically, Texas experienced a 19.7 percent decrease in its aggregate violent crim ate in ontro the ten years following its RTC law (between 1996 and 2006), while the state’s syntheti experienced a larger 31.0 percent decline. This counterfactual would not be apparent t esidents of the state or to law enforcement officials, but our results suggest that Texas’s RTC la mposed a large social cost on the state. imposed a large social cost on the state.
erceived favorable experience with the law, we see why it can be quite easy to draw the inaccurat ausal inference that Texas’ crime decline was facilitated by its RTC law. The public may perceiv he fal ing crime rate post-1996 (the solid black line) but our analysis suggests that Texas woul lave experienced a more sizable violent crime decline if it had not passed an RTC law (the dotte ine). More specifically, Texas experienced a 19.7 percent decrease in its aggregate violent crim ate in ontro the ten years following its RTC law (between 1996 and 2006), while the state’s syntheti experienced a larger 31.0 percent decline. This counterfactual would not be apparent t esidents of the state or to law enforcement officials, but our results suggest that Texas’s RTC la mposed a large social cost on the state. imposed a large social cost on the state.
2. State-Specific Estimates Across all RTC States 3ecause we are projecting the violent crime experience of the synthetic control over a ten-yea: yeriod, there will undoubtedly be a deviation from the “true” counterfactual and our estimatec -ounterfactual. If we were only estimating the impact of a legal change for a single state, we vould have an estimate marred by this purely stochastic aspect of changing crime. Since we ire estimating an average effect across a large number of states, the stochastic variation will be liminished as the over-estimates and under-estimates will tend to wash out in our mean treatmen »stimates. Figure 5 shows the synthetic control estimates on violent crime for all 31 states for whicl ve have ten years of post-passage data. For 23 of the 31 states adopting RTC laws, the increase ir Because we are projecting the violent crime experience of the synthetic control over a ten-year diminished as the over-estimates and under-estimates will tend to wash out in our mean treatment
2. State-Specific Estimates Across all RTC States 3ecause we are projecting the violent crime experience of the synthetic control over a ten-yea: yeriod, there will undoubtedly be a deviation from the “true” counterfactual and our estimatec -ounterfactual. If we were only estimating the impact of a legal change for a single state, we vould have an estimate marred by this purely stochastic aspect of changing crime. Since we ire estimating an average effect across a large number of states, the stochastic variation will be liminished as the over-estimates and under-estimates will tend to wash out in our mean treatmen »stimates. Figure 5 shows the synthetic control estimates on violent crime for all 31 states for whicl ve have ten years of post-passage data. For 23 of the 31 states adopting RTC laws, the increase ir Because we are projecting the violent crime experience of the synthetic control over a ten-year diminished as the over-estimates and under-estimates will tend to wash out in our mean treatment
V. Aggregation Analysis Using Synthetic (“‘ontroals
V. Aggregation Analysis Using Synthetic (“‘ontroals
Table 5: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, 1977-2014 We begin our analysis of the aggregated synthetic control results using predictors derived from the DAW specification. Table 5 shows our results on the full sample examining violent crime.™’ Our estimates of the normalized average treatment effect percentage (TEP) suggest that states that passed RTC laws experienced more deleterious changes in violent criminal activity than thei synthetic controls in the ten years after adoption. On average, treatment states had aggregate violent crime rates that were almost seven percent higher than their synthetic controls five years after passage and around 14 percent higher ten years after passage. Table 5 suggests that the longet the RTC law is in effect (up to the tenth year that we analyze), the greater the cost in terms of increased violent crime. the RTC law is in effect (up to the tenth year that we analyze), the greater the cost in terms of
Table 5: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, 1977-2014 We begin our analysis of the aggregated synthetic control results using predictors derived from the DAW specification. Table 5 shows our results on the full sample examining violent crime.™’ Our estimates of the normalized average treatment effect percentage (TEP) suggest that states that passed RTC laws experienced more deleterious changes in violent criminal activity than thei synthetic controls in the ten years after adoption. On average, treatment states had aggregate violent crime rates that were almost seven percent higher than their synthetic controls five years after passage and around 14 percent higher ten years after passage. Table 5 suggests that the longet the RTC law is in effect (up to the tenth year that we analyze), the greater the cost in terms of increased violent crime. the RTC law is in effect (up to the tenth year that we analyze), the greater the cost in terms of
Table 6: The Impact of RTC Laws on the Violent Crime Rate, LM covariates, Full Sample. 1977-2014 D. The Contributions of Donor States to the Synthetic Control Estimates: Evaluating Robustness
Table 6: The Impact of RTC Laws on the Violent Crime Rate, LM covariates, Full Sample. 1977-2014 D. The Contributions of Donor States to the Synthetic Control Estimates: Evaluating Robustness
The impact of gun ownership on the increase in violent crime due to RTC laws (synthetic control estimates, 1977-2014) F. The Murder and Property Crime Assessments with Synthetic Controls
The impact of gun ownership on the increase in violent crime due to RTC laws (synthetic control estimates, 1977-2014) F. The Murder and Property Crime Assessments with Synthetic Controls
Prior to 1995, the number of concealed carry permits remained low because of the requirement
Prior to 1995, the number of concealed carry permits remained low because of the requirement
Table B1: Panel Data Violent Crime Coefficients using DAW and LM models, State and Year Fixed Effects
Table B1: Panel Data Violent Crime Coefficients using DAW and LM models, State and Year Fixed Effects
Estimations include year and state fixed effects and are weighted by state population. Coefficients on demographic variables and the constant omitted. Robust standard errors (clustered at the state level) are provided next to point estimates in parentheses. The crime data is from the Uniform Crime Reports (UCR). * p <.1, ** p < .05, *** p < .01. All figures reported in percentage terms. The DAW model is run on data from 1979-2014, the BC model from 1978-2014, the LM model from 1977-2014, and the MM model (without the crack cocaine index) from 1979-2014.
Estimations include year and state fixed effects and are weighted by state population. Coefficients on demographic variables and the constant omitted. Robust standard errors (clustered at the state level) are provided next to point estimates in parentheses. The crime data is from the Uniform Crime Reports (UCR). * p <.1, ** p < .05, *** p < .01. All figures reported in percentage terms. The DAW model is run on data from 1979-2014, the BC model from 1978-2014, the LM model from 1977-2014, and the MM model (without the crack cocaine index) from 1979-2014.
Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate. The impact of RTC laws on violent crime, DAW model, 2000-2014 Figure Cl
Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate. The impact of RTC laws on violent crime, DAW model, 2000-2014 Figure Cl
All models include year and state fixed effects, and the OLS estimates are weighted by state population. Robust standard errors (clustered at the state level) are provided next to point estimates in parentheses. In Column 2 we present IRRs estimated using negative binomial regression, where population is included as a control variable. The following 11 states adopted RTC Laws during the period of consideration: CO (2003), IA (2011), IL (2014), KS (2007), MI (2001), MN (2003), MO (2004), NE (2007), NM (2004), OH (2004), and WI (2011). *p<.1, ** p< .05, *** p < .01. All figures reported in percentage terms.
All models include year and state fixed effects, and the OLS estimates are weighted by state population. Robust standard errors (clustered at the state level) are provided next to point estimates in parentheses. In Column 2 we present IRRs estimated using negative binomial regression, where population is included as a control variable. The following 11 states adopted RTC Laws during the period of consideration: CO (2003), IA (2011), IL (2014), KS (2007), MI (2001), MN (2003), MO (2004), NE (2007), NM (2004), OH (2004), and WI (2011). *p<.1, ** p< .05, *** p < .01. All figures reported in percentage terms.
The impact of RTC laws on murder, DAW model, 2000-2014 Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on murder, DAW model, 2000-2014 Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on violent crime, LM model, 1977-2014 Note: We regress crime on dummies for pre— and post-passage years and LM covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on violent crime, LM model, 1977-2014 Note: We regress crime on dummies for pre— and post-passage years and LM covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on property crime, DAW model, 1979-2014 Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on property crime, DAW model, 1979-2014 Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on property crime, LM model, 1977-2014 Note: We regress crime on dummies for pre— and post-passage years and LM covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on property crime, LM model, 1977-2014 Note: We regress crime on dummies for pre— and post-passage years and LM covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on murder, DAW model, 1979-2014 Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on murder, DAW model, 1979-2014 Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on murder, LM model, 1977-2014 Note: We regress crime on dummies for pre— and post-passage years and LM covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on murder, LM model, 1977-2014 Note: We regress crime on dummies for pre— and post-passage years and LM covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on murder, DAW model, negative binomial model, 1979-2014 Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
The impact of RTC laws on murder, DAW model, negative binomial model, 1979-2014 Note: We regress crime on dummies for pre— and post-passage years and DAW covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate.
Note: We regress crime on dummies for pre— and post-passage years and LM covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate. The impact of RTC laws on murder, LM model, negative binomial model,1977-2014
Note: We regress crime on dummies for pre— and post-passage years and LM covariates. Reference year is year before adoption and adoption year is first year with RTC in place at any time, meaning that in states that adopt after January 1st, this will capture only a partial effect of RTC laws. We display the 95 percent confidence interval for each estimate using cluster—-robust standard errors and show the number of states that contribute to each estimate. The impact of RTC laws on murder, LM model, negative binomial model,1977-2014
Frequency of potential donor states to appear as synthetic controls in violent crime estimates
Frequency of potential donor states to appear as synthetic controls in violent crime estimates
The graph shows the overall synthetic control estimate of the impact of RTC laws on violent crime ten years after adoption when barring individual states from inclusion in the synthetic control. (The horizontal line shows the estimate when no states are barred.) The states are arranged in declining order of population-weighted average contribution to synthetic controls (see Appendix Figure D8), from a high of 18.1 percent for California to a low of 0.2 percent for Minnesota.
The graph shows the overall synthetic control estimate of the impact of RTC laws on violent crime ten years after adoption when barring individual states from inclusion in the synthetic control. (The horizontal line shows the estimate when no states are barred.) The states are arranged in declining order of population-weighted average contribution to synthetic controls (see Appendix Figure D8), from a high of 18.1 percent for California to a low of 0.2 percent for Minnesota.
Table E1: The Impact of RTC Laws on the Murder Rate, DAW covariates, Full Sample. 1977-2014 Table E2: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, Full Sample, 1977-2014 in the murder rate from RTC adoption range from 11-14 percent, and are statistically significant at
Table E1: The Impact of RTC Laws on the Murder Rate, DAW covariates, Full Sample. 1977-2014 Table E2: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, Full Sample, 1977-2014 in the murder rate from RTC adoption range from 11-14 percent, and are statistically significant at
Table E4: The Impact of RTC Laws on the Property Crime Rate, LM covariates, Full Sam- ple, 1977-2014
Table E4: The Impact of RTC Laws on the Property Crime Rate, LM covariates, Full Sam- ple, 1977-2014
Table F1: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, Full Sample 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F1: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, Full Sample 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F2: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, < 2x Av- erage Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F2: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, < 2x Av- erage Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F3: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, < 1x Av- erage Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F3: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, < 1x Av- erage Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F4: The Impact of RTC Laws on the Murder Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F4: The Impact of RTC Laws on the Murder Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F5: The Impact of RTC Laws on the Murder Rate, DAW covariates, < 2x Average Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F5: The Impact of RTC Laws on the Murder Rate, DAW covariates, < 2x Average Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F6: The Impact of RTC Laws on the Murder Rate, DAW covariates, < 1x Average Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F6: The Impact of RTC Laws on the Murder Rate, DAW covariates, < 1x Average Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F7: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F7: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F8: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, < 2) Average Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Yea Crime Differential
Table F8: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, < 2) Average Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Yea Crime Differential
Table F9: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, < 1x Average Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Table F9: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, < 1x Average Coefficient of Variation of the RMSPE, 1977-2014, No Subtraction of Adoption Year Crime Differential
Fisures 11-133 Appendix I: Synthetic Control Graphs Estimating Impact of RTC Laws On Violent Crime Using the DAW Model®> Laws On Violent Crime Using the DAW Model® Appendix I: Synthetic Control Graphs Estimating Impact of RTC 85Recall that each state’s effective year of passage is defined as the first year in which an RTC law was in effect for the majority of that year.
Fisures 11-133 Appendix I: Synthetic Control Graphs Estimating Impact of RTC Laws On Violent Crime Using the DAW Model®> Laws On Violent Crime Using the DAW Model® Appendix I: Synthetic Control Graphs Estimating Impact of RTC 85Recall that each state’s effective year of passage is defined as the first year in which an RTC law was in effect for the majority of that year.
Notes: After getting a measure of fit for each state, an unweighted average is taken to arrive at a single measure of fit.Training Period from 1977 through RTC year - 6; Validation Period from RTC year - 5 through RTC vear - 1 Table K1: Define Fit Using Mean Squared Error
Notes: After getting a measure of fit for each state, an unweighted average is taken to arrive at a single measure of fit.Training Period from 1977 through RTC year - 6; Validation Period from RTC year - 5 through RTC vear - 1 Table K1: Define Fit Using Mean Squared Error
Notes: After getting a measure of fit for each state, an unweighted average is taken to arrive at a single measure of fit. Training Period from 1977 through RTC year - 6; Validation Period from RTC year - 5 through RTC year - 1 Table K2: Define Fit Using Mean Absolute Difference
Notes: After getting a measure of fit for each state, an unweighted average is taken to arrive at a single measure of fit. Training Period from 1977 through RTC year - 6; Validation Period from RTC year - 5 through RTC year - 1 Table K2: Define Fit Using Mean Absolute Difference
Notes: After getting a measure of fit for each state, an unweighted average is taken to arrive at a single measure of fit. Training Period from 1977 through RTC year - 6; Validation Period from RTC year - 5 through RTC year - |
Notes: After getting a measure of fit for each state, an unweighted average is taken to arrive at a single measure of fit. Training Period from 1977 through RTC year - 6; Validation Period from RTC year - 5 through RTC year - |
Notes: After getting a measure of fit for each state, a population weighted average is taken to arrive at a single measure of fit. Training Period from 1977 through RTC year - 6 Validation Period from RTC year - 5 through RTC year - 1. Population from first year of relevant period is used. Table K4: Define Fit Using Mean Squared Error
Notes: After getting a measure of fit for each state, a population weighted average is taken to arrive at a single measure of fit. Training Period from 1977 through RTC year - 6 Validation Period from RTC year - 5 through RTC year - 1. Population from first year of relevant period is used. Table K4: Define Fit Using Mean Squared Error
Notes: After getting a measure of fit for each state, a population weighted average is taken to arrive at a single measure of fit.Training Period from 1977 through RTC year - | Validation Period from RTC year - 5 through RTC year - 1. Population from first year of relevant period is used. Table K5: Define Fit Using Mean Absolute Difference
Notes: After getting a measure of fit for each state, a population weighted average is taken to arrive at a single measure of fit.Training Period from 1977 through RTC year - | Validation Period from RTC year - 5 through RTC year - 1. Population from first year of relevant period is used. Table K5: Define Fit Using Mean Absolute Difference
Notes: After getting a measure of fit for each state, a population weighted average is taken to arrive at a single measure of fit.Training Period from 1977 through RTC year - 6; Validation Period from RTC year - 5 through RTC year - 1. Population from first year of relevant period is used.
Notes: After getting a measure of fit for each state, a population weighted average is taken to arrive at a single measure of fit.Training Period from 1977 through RTC year - 6; Validation Period from RTC year - 5 through RTC year - 1. Population from first year of relevant period is used.
Notes: We take the average of the absolute percentage difference in economic predictors between Treatment and Synthetic Control states using five lag specifications. The values reported are the ratio of this statistic for each lag specification to the average of this statistic for all five lag choices. Age groups represent the percent of population that is white male, black male or other male in two age brackets (15-19 and 20-39).
Notes: We take the average of the absolute percentage difference in economic predictors between Treatment and Synthetic Control states using five lag specifications. The values reported are the ratio of this statistic for each lag specification to the average of this statistic for all five lag choices. Age groups represent the percent of population that is white male, black male or other male in two age brackets (15-19 and 20-39).
Standard errors in parentheses Column numbers indicate post-passage year under consideration; N = number of states in sample Dependent variable is the difference between the percentage difference in the violent crime rate in treatment and synthetic control states at given post-treatment interval and at time of the treatment Results reported for the constant term resulting from this regression States in group: AK AR AZ CO FL GA ID KS KY LA ME MI MN MO MS MT NC ND NE NM NV OH OK OR PA SC SD TN TX UT VA W‘ *p<0.10, * p< 0.05, *** p< 0.01 Table K8: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, Variou: Lag specifications, Full Sample, 1977-2014
Standard errors in parentheses Column numbers indicate post-passage year under consideration; N = number of states in sample Dependent variable is the difference between the percentage difference in the violent crime rate in treatment and synthetic control states at given post-treatment interval and at time of the treatment Results reported for the constant term resulting from this regression States in group: AK AR AZ CO FL GA ID KS KY LA ME MI MN MO MS MT NC ND NE NM NV OH OK OR PA SC SD TN TX UT VA W‘ *p<0.10, * p< 0.05, *** p< 0.01 Table K8: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, Variou: Lag specifications, Full Sample, 1977-2014
Table L1: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, Full Sample, 1977-2014
Table L1: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, Full Sample, 1977-2014
Table L2: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction
Table L2: The Impact of RTC Laws on the Violent Crime Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction
Table L4: The Impact of RTC Laws on the Murder Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction
Table L4: The Impact of RTC Laws on the Murder Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction
Table L5: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, Full Sample, 1977-2014
Table L5: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, Full Sample, 1977-2014
Table L6: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction
Table L6: The Impact of RTC Laws on the Property Crime Rate, DAW covariates, Full Sample, 1977-2014, No Subtraction
Related topics:
LawPolitical ScienceEmpirical legal studies
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved