Partial effects in non-linear panel data models with correlated random effects.

Nonlinearity and heterogeneity are known to cause difficulties in estimating and interpreting partial effects. This paper provides a systematic characterization of the various partial effects in nonlinear panel data models that might be of interest to empirical researchers. The interpretation of the partial effects depends upon (i) whether the distribution of unobserved heterogeneity is treated as fixed or allowed to vary with covariates, and (ii) whether one is interested in particular covariate values or an average over such values. The characterization covers partial-effects concepts already in the literature but also includes new concepts for partial effects. A simple panel probit design highlights that the different partial effects can be quantitatively very different.

Interval censored regression with fixed effects.

This paper considers identification and estimation of a fixed-effects model with an interval-censored dependent variable. In each time period, the researcher observes the interval (with known endpoints) in which the dependent variable lies but not the value of the dependent variable itself. Two versions of the model are considered: a parametric model with logistic errors and a semiparametric model with errors having an unspecified distribution. In both cases, the error disturbances can be heteroskedastic over cross-sectional units as long as they are stationary within a cross-sectional unit; the semiparametric model also allows for serial correlation of the error disturbances. A conditional-logit-type composite likelihood estimator is proposed for the logistic fixed-effects model, and a composite maximum-score-type estimator is proposed for the semiparametric model. In general, the scale of the coefficient parameters is identified by these estimators, meaning that the causal effects of interest are estimated directly in cases where the latent dependent variable is of primary interest (e.g., pure data-coding situations). Monte Carlo simulations and an empirical application to birthweight outcomes illustrate the performance of the parametric estimator.

Missing dependent variables in fixed-effects models.

This paper considers estimation of linear fixed-effects models in which the dependent variable may be missing. For cross-sectional units with dependent variables missing, use of covariate information from all time periods can provide efficiency gains relative to complete-data methods. A classical minimum distance (CMD) estimator based upon Chamberlain (1982, 1984), which is consistent under a missing-at-random (MAR) type assumption, is proposed for the static fixed-effects model. In certain circumstances, it is shown that "within" variation in the dependent variable is not even required for identification of the model parameters. The CMD estimation approach is extended to the case of (autoregressive) fixed-effects models with lagged dependent variables. Monte Carlo simulations investigate the performance of the CMD approach relative to existing methods. Extensions to models with sequential exogeneity and missing covariates are also discussed.

Examination of universal purchase programs as a driver of vaccine uptake among US States, 1995-2014.

BACKGROUND: Immunization against numerous potentially life-threatening illnesses has been a great public health achievement. In the United States, the Vaccines for Children (VFC) program has provided vaccines to uninsured and underinsured children since the early 1990s, increasing vaccination rates. In recent years, some states have adopted Universal Purchase (UP) programs with the stated aim of further increasing vaccination rates. Under UP programs, states also purchase vaccines for privately-insured children at federally-contracted VFC prices and bill private health insurers for the vaccines through assessments. METHODS: In this study, we estimated the effect of UP adoption in a state on children's vaccination rates using state-level and individual-level data from the 1995-2014 National Immunization Survey. For the state-level analysis, we performed ordinary least squares regression to estimate the state's vaccination rate as a function of whether the state had UP in the given year, state demographic characteristics, other vaccination policies, state fixed effects, and a time trend. For the individual analysis, we performed logistic regression to estimate a child's likelihood of being vaccinated as a function of whether the state had UP in the given year, the child's demographic characteristics, state characteristics and vaccine policies, state fixed effects, and a time trend. We performed separate regressions for each of nine recommended vaccines, as well as composite measures on whether a child was up-to-date on all required vaccines. RESULTS: In the both the state-level and individual-level analyses, we found UP had no significant (p < 0.10) effect on any of the vaccines or composite measures in our base case specifications. Results were similar in alternative specifications. CONCLUSIONS: We hypothesize that UP was ineffective in increasing vaccination rates. Policymakers seeking to increase vaccination rates would do well to consider other policies such as addressing provider practice issues and vaccine hesitancy.

Effectiveness of state-level vaccination mandates: evidence from the varicella vaccine.

This paper utilizes longitudinal data on varicella (chickenpox) immunizations in order to estimate the causal effects of state-level school-entry and daycare-entry immunization mandates within the United States. We find significant causal effects of mandates upon vaccination rates among preschool children aged 19-35 months; these effects appear in the year of mandate adoption, peak two years after adoption, and show a minimal difference from the aggregate trend about six years after adoption. For a mandate enacted in 2000, the model and estimates imply that roughly 20% of the short-run increase in state-level immunization rates was caused by the mandate introduction. We find no evidence of differential effects for different socioeconomic groups. Combined with previous cost-benefit analyses of the varicella vaccine, the estimates suggest that state-level mandates have been effective from an economic standpoint.

Temper, temperature, and temptation: heat-related retaliation in baseball.

In this study, we analyzed data from 57,293 Major League Baseball games to test whether high temperatures interact with provocation to increase the likelihood that batters will be hit by a pitch. Controlling for a number of other variables, we conducted analyses showing that the probability of a pitcher hitting a batter increases sharply at high temperatures when more of the pitcher's teammates have been hit by the opposing team earlier in the game. We suggest that high temperatures increase retaliation by increasing hostile attributions when teammates are hit by a pitch and by lowering inhibitions against retaliation.

Trends and determinants of second-pregnancy smoking among young-adult mothers who smoked during their first pregnancy.

We examined the trends and determinants of second-pregnancy smoking behavior among Michigan young-adult mothers who smoked during their first pregnancy (N = 14,731). Maternally linked Michigan birth certificate data (1989-2004) were used to identify first and second pregnancies for young-adult mothers. ZIP code data were merged with census data to proxy for median household income and population size at the ZIP code level. For first-pregnancy smokers, linear and logistic regressions were used to analyze (a) the determinants of second-pregnancy smoking and (b) the association of birth outcomes with second-pregnancy smoking. Among young adults who smoked during their first pregnancy, second-pregnancy smoking was (a) inversely related to income and population, (b) less prevalent after 1997, (c) more likely after a low-birthweight first birth, and (d) less likely after a premature first birth. Despite recent declines in second-pregnancy smoking rates among young adults who smoked during their first pregnancy, these rates remain disturbingly high (more than 70%). Future smoking cessation efforts should target mothers who have lower incomes, live in less-populated areas, and have previously had low-birthweight babies. The results suggest that higher cigarettes taxes would also further reduce recurrent smoking.