![Question: 5. 30pts] For this problem you will use data fromC and Macstas 2009 paper uity design Does Medica...](http://d2vlcm61l7u1fs.cloudfront.net/media%2F364%2F36453aa7-bf84-45d8-ba6b-20b600677d41%2FphpVvs8ro.png)
![Question: 5. 30pts] For this problem you will use data fromC and Macstas 2009 paper uity design Does Medica...](http://d2vlcm61l7u1fs.cloudfront.net/media%2F69e%2F69e2aab0-0446-468f-93e1-17eae8d8068e%2FphpfRtAwF.png)
![Question: 5. 30pts] For this problem you will use data fromC and Macstas 2009 paper uity design Does Medica...](http://d2vlcm61l7u1fs.cloudfront.net/media%2Ff90%2Ff9087d43-e51d-4f20-a6ae-acb9c83cdd18%2FphpkHLc0c.png)
see more
Show transcribed image text 5. 30pts] For this problem you will use data fromC and Macstas 2009 paper uity design Does Medicare Save Lives?". This paper specifically used a regression di (RD) to evaluate whether health insurance (treatment) affects overall health. Since at 65 all legal residents in the U.S. are eligible for Medicare, subsidized health insurance for the elderly, we can compare health outcomes of people on cither side of this cut-off to potentially get a casual effect of health insurance on health. You will be replicating their study in a more simplistic way for this question. The data you will use is called card et_al.dta and can be found on blackboard. It holds health information, insurance status, and age for a random sample of individuals between the ages of 55-74. Please load it into Stata and create these two variables with the code below: generate age65 ag generate stan_age-age_gtr-260 e-gtr>259 The first variable (age65) is an indicator variable that the person is eligible for Medicare. The second variable rescales age in quarters so that stan_age 0 corresponds to the 1" quarter of Medicare eligibility while negative values correspond to the number months until a person becomes eligible for Medicare and positive values indicate how long a person has been eligible for Medicare. a) Graph the variable prop_insured (variable which indicates the proportion of those insured by age) against stan_age and include this figure in your assignment. To do create the graph use this code: twoway scatter prop insured stan age What do you notice about proportion people insured at the cut-off for Medicare eligibility (stan_age-0)? b) To estimate the relationship between health insurance and health outcomes we will be using fuzzy RD. Please explain why we are using fuzzy RD instead of sharp RD (hint use graph from part (a) to help justify). c Fuzzy RD is basically Instrumental Variables. Let's run the first stage: the effect of Medicare eligibility on insurance. We will run the first-stage two ways: once without the running variable (stan_age) and once with the running variable (stan age). Use the code below to run the different models for first stage and include output in your
5. 30pts] For this problem you will use data fromC and Macstas 2009 paper uity design Does Medicare Save Lives?". This paper specifically used a regression di (RD) to evaluate whether health insurance (treatment) affects overall health. Since at 65 all legal residents in the U.S. are eligible for Medicare, subsidized health insurance for the elderly, we can compare health outcomes of people on cither side of this cut-off to potentially get a casual effect of health insurance on health. You will be replicating their study in a more simplistic way for this question. The data you will use is called card et_al.dta and can be found on blackboard. It holds health information, insurance status, and age for a random sample of individuals between the ages of 55-74. Please load it into Stata and create these two variables with the code below: generate age65 ag generate stan_age-age_gtr-260 e-gtr>259 The first variable (age65) is an indicator variable that the person is eligible for Medicare. The second variable rescales age in quarters so that stan_age 0 corresponds to the 1" quarter of Medicare eligibility while negative values correspond to the number months until a person becomes eligible for Medicare and positive values indicate how long a person has been eligible for Medicare. a) Graph the variable prop_insured (variable which indicates the proportion of those insured by age) against stan_age and include this figure in your assignment. To do create the graph use this code: twoway scatter prop insured stan age What do you notice about proportion people insured at the cut-off for Medicare eligibility (stan_age-0)? b) To estimate the relationship between health insurance and health outcomes we will be using fuzzy RD. Please explain why we are using fuzzy RD instead of sharp RD (hint use graph from part (a) to help justify). c Fuzzy RD is basically Instrumental Variables. Let's run the first stage: the effect of Medicare eligibility on insurance. We will run the first-stage two ways: once without the running variable (stan_age) and once with the running variable (stan age). Use the code below to run the different models for first stage and include output in your
