Econometrics

A proper regression   is a major portion of the grade for the assignment. It needs a discussion of the data you collected, a discussion of the strengths and weaknesses of the data, the regression results (coefficients, t stats, R square, F stat, etc), diagnostics (multicollinearity, heteroscedasticity, autocorrelation), and analysis of the results.

Plus complete all the rubric requirements 

ECO 625: Milestone Two Guidelines and Rubric

Overview: The final project for this course is designed to guide you through the process of applying the key components of econometrics. In the previous
econometrics course (ECO 620), you practiced using regression analysis, which is useful but has some serious limitations and is not appropriate in many settings.
As an economist, you will often be asked to analyze data that has serious nonlinearity or is not normally distributed, and you must be able to rise to this
challenge. In order to practice the skills involved, you will create a final project in which you will prepare an econometric analysis of a business, policy, or
economic issue of your choice, utilizing advanced methods studied in this course (forecasting, nonparametric analysis, maximum likelihood, and so on).

You should choose a topic of personal or professional interest to you. You may revisit the topic you selected in ECO 620 or choose a new topic. A literature review
of peer-reviewed publications will be necessary in order to familiarize yourself with possible ways to address the issue empirically, using advanced econometric
methods. You will build an empirical model to analyze the issue and use the data to conduct econometric analysis. You will then explain your findings to both
technical and nontechnical audiences.

Prompt: Submit a two- to three-page paper that addresses the following questions: What are the quantitative implications and actionable insights of the chosen
business, policy, or economic issue? Also, what is the benefit of using advanced econometric techniques relative to simpler ones, such as regression?

Specifically, the following critical elements must be addressed:

I. Data
A. What limitations could your data impose on the choice of empirical method? What are the effects of these limitations?

II. Empirical Approach
A. What empirical method(s) do you propose and why? Why is this method most appropriate and preferable to simpler models, such as regression?
B. What are the limitations of your proposed empirical method? What are possible alternative advanced methods that can be used? Be sure to

address causes for these limitations, such as problems with the data or with interactions between variables.
C. What is your model specification? What functional form of data do you use and why?

III. Results and Robustness
A. What are your preliminary results from baseline estimations in the statistical software? What is their relationship with the original hypothesis

and research question? Is there a difference between the baseline estimation and the ultimate choice of the model? Discuss.
B. Which violations of the chosen model’s assumptions do you anticipate? Which diagnostic tests do you employ to check for presence of

violations of model assumptions?
C. What are your secondary results from your test run through the statistical software? Interpret these test results and determine the presence or

absence of problems.
D. Are your results affected by the corrections? Are they affected in a significant way (e.g., change of sign of coefficients or change in statistical

significance)?

Rubric

Guidelines for Submission: Your submission should be two to three pages in length (not including title page and references) and should use double spacing,
12-point Times New Roman font, one-inch margins, and citations in APA format.

Critical Elements Exemplary (100%) Proficient (90%) Needs Improvement (70%) Not Evident (0%) Value

Data: Limitations

Meets “Proficient” criteria and
supports discussion with
scholarly research

Identifies limitations of the data
structure on the choice of
empirical methods and details
the effects of these limitations

Identifies limitations of the data
structure on the choice of
empirical methods but does not
detail the effects of these
limitations, or discussion misses
key limitations

Does not identify limitations of
the data structure on the choice
of empirical methods

10

Empirical Approach:
Method

Meets “Proficient” criteria and
explains the reasons for the
superiority of the chosen
method(s) over other methods

Proposes advanced empirical
method(s) to use in the study
and clearly defends method(s),
stating the considerations
leading to the choice, including
why simpler models were not
employed

Proposes advanced empirical
method(s) to use in the study
and defends these method(s)
but does not address
considerations leading to the
choice, or defense of proposed
methods is lacking in detail

Does not propose and defend
advanced empirical method(s)
to use in the study

10

Empirical Approach:
Limitations

Meets “Proficient” criteria and
supports discussion with
scholarly research

Identifies limitations of the
proposed model, delineating
between issues with the data
and interactions between
variables, and addresses the
appropriateness of alternative
methods/models

Identifies limitations of the
proposed model but does not
delineate between issues with
the data and interactions
between variables or does not
address the appropriateness of
alternative methods/models

Does not identify limitations of
the proposed model

10

Empirical Approach:
Specification

Meets “Proficient” criteria, and
justification is supported
through research

Outlines the empirical approach
in a clear manner, providing
specifications in the form of an
equation and justifying the
transformation

Outlines the empirical approach
but does not provide
specifications in the form of an
equation or does not justify the
transformation

Does not outline the empirical
approach

10

Results and
Robustness:

Preliminary Results

Meets “Proficient” criteria and
discusses any counterintuition in
the results or possible
inconsistencies

Utilizes statistical software to
produce an initial report of
results, details those results, and
determines the relationship of
the results to the hypothesis and
research question, disclosing
differences in baseline
estimation and ultimate choice
of model

Utilizes statistical software to
produce an initial report of
results but does not present
those results beyond simple
statements of fact or does not
determine the relationship of
the results to the hypothesis and
research question

Does not utilize statistical
software to produce an initial
report of results

1

5

Results and
Robustness:
Violations

Meets “Proficient” criteria, and
examination addresses the
effects that identified ordinary
least squares violations would
have on results

Identifies violations and chooses
appropriate statistical tests
based on initial examination of
results

Identifies violations but does not
choose appropriate statistical
tests to run or employ a
complete battery of tests

Misinterprets results of the tests
or uses inappropriate tests

15
Results and
Robustness:

Secondary Results

Meets “Proficient” criteria, and
interpretation is highly specific
and detailed

Conducts, interprets, and
reports results from secondary
tests correctly and determines
the presence or absence of
problem(s)

Conducts and reports results
from secondary tests, but
interpretation of test results is
limited/incorrect, or the
presence/absence of problem(s)
is not determined

Does not conduct or report
results from secondary tests

15

Results and
Robustness: Affected

Meets “Proficient” criteria and
provides illustrative examples of
impact of changes on overall
conclusions

Determines the effects of
corrective actions by noting
changes (or lack thereof) and
the magnitude in coefficients of
these changes in statistical
significance of key independent
variables

Determines the effects of
corrective actions but does not
note changes (or lack thereof) or
does not note the magnitude in
coefficients of these changes in
statistical significance of key
independent variables, or the
determination of effects is
flawed

Does not determine the effects
of corrective actions

10

Articulation of
Response

Submission is free of errors
related to citations, grammar,
spelling, syntax, and
organization and is presented in
a professional and easy to read
format

Submission has no major errors
related to citations, grammar,
spelling, syntax, or organization

Submission has major errors
related to citations, grammar,
spelling, syntax, or organization
that negatively impact
readability and articulation of
main ideas

Submission has critical errors
related to citations, grammar,
spelling, syntax, or organization
that prevent understanding of
ideas

5

Earned Total 100%

Introduction

Several economic types of research have demonstrated that there is a strong positive correlation between years of schooling and health

.

However, the main question centered in this study is the relationship that exists between education and Health (Buckles, et al.2013). This paper will employ several changes that have been made in education and health studies to test the hypothesis that there is a causal relationship between education and health. Results from this study suggest that there is a causal relation ranging from more schooling to better health, which is more significant than the standards regression suggestions

Description

Public intellectuals and policymakers usually emphasize the essence of education. They argue that education results in expanded job opportunities and higher expected earnings. However, there may be other essential benefits of education, which have not been understood appropriately. Recent economic literature reviews on the effects of education on the health of a population found out that there is substantial evidence that links education not only to increase earning potential of an individual but also to reduce criminal behavior. This is also related to increased voting as well as democratic participation and improved health outcomes. Given the fact that education is a crucial multifaceted component that affects health; the research composed in this paper has education and health policy makers, as its targets audiences due to the multiple causative relationships between the two variables. The ability of policymakers and the governments to understand the Education- Health relationship would help them whenever deciding on whether to invest more in education or healthcare.

.

Literature Review

With the current empirical economics, hypotheses usually go either way, depending on the economist’s perspective. One might assume that better education leads to better health or better health lead to a better education. Or maybe the fact that education brings more income thus betters health; versus better health helping individuals become more educated. But one thing that we could all agree on is the fact that education correlates with health. Education is one of the major social factors that most economic researchers have cited that is linked to longer lifespans in every country where it has been studied. For example; according to the CDC: for every 100,000 deaths amongst non-high school graduate American males aged between 25 to 64 years old, the mortality rate was 655.2; for the males within the same age group but with high-school diplomas, the mortality rate is 600.9. Whereas; the mortality rate for those with college education or higher given the same parameters was 238.9(Martinek, 2017). Such results are a pure reflection of the fact that the more educated people are, the more likely they are better informed thus making better health choices.

Alternatively, health in young adulthood and childhood years may affect the ability of an individual to receive an education. Educational attainment affects income as well as other aspects of people’s lives, which facilitate access to medical and preventive care. Individuals that are more educated can practice safer behaviors such as deciding not to smoke, often using a seatbelt, take vaccinations, and even exercising regularly more often than an individual that is less educated.

Despite the existence of an empirical relationship between the two constraints, it has been hard to establish the existence of a causal relationship between health and education. This is because of the dual causality issue. The underlying rapport can be rooted from the concept of the disputation that education brings more income, which in return allows better access to better healthcare. But the difference in the income levels only accounts for 20% of the effect that advanced education has on health behaviors (Martinek, 2017).

Additionally, other unseen variables i.e. individual-level behaviors such as patience that make a person to more likely invest in both education and in the long-term health may contribute to the enhanced relationship between education and health. It is therefore hard to find out what way the causation goes and the mechanisms that encourage these effects. Most of the hard empirical analysis addresses the issue of causality emphasis on compulsory educational law changes that were made in 1960. It’s for this reason that I’m testing of the hypothesis that education has a crucial social effect on health.

While in search of supporting evidence, we will comb through similar studies that have tried to detect causal effects with robust assessment techniques.

Data

Large scale demographic and improvement studies likewise recognize a critical positive relationship between extending mass education and the health of people. However, numerous other studies particularly the most punctual ones, did not control different health components such as monetary advancement that may be connected to better health. As an outcome, it is constantly recommended that the relationship between extending education and population health is unauthentic due to the effect of general social improvement that includes financial development and the nature of human services. Subsequently, related variables of innovation are frequently expected to prepare for both mass education and more health population. While holding other variables constant; except mortality rate, we could proof the effect of education on health using the data from the National Longitudinal Mortality Study (NLMS), where one more year of education increases life expectancy by 0.18 years; with the discount rate at 3% or 0.6 of a year while disregarding the discount: If we were to assume that each health year is worth conservatively at $75000; This would directly translates into about $13,500 – $44,000 in present value. With such rough estimates we clearly see that the health returns to education upsurge the total returns to education by at least 15% on the lower end and 55 on the upper end.

Works Cited
Martinek, C. (2017, 12 5). Which Came First-Better Education or Better Health? Retrieved 12 15, 2017, from Which Came First: Better Education or Better Health? | St. Louis Fed: http://www.stlouisfed.org/publications/re/articles/?id=2092
Buckles, Kasey, Andreas Hagemann, Ofer Malamud, Melinda S. Morrill, and Abigail K.
Wozniak. 2013. “The Effect of College Education on Health.” Working Paper 19222.
Goesling, B. (2007). “The Rising Significance of Education for Health?” Social Forces 85 (4):
1621–44.
Grossman, M. 2015. “The Relationship between Health and Schooling: What’s New?”
Working Paper 21609. National Bureau of Economic Research.
http://www.nber.org/papers/w21609.
Institute of Education Sciences. 2015. “Digest of Education Statistics, 2013.” Accessed
http://nces.ed.gov/programs/digest/d13/tables/dt13_303.40.asp.
Lleras-Muney, A. (2006). “The Relationship between Education and Adult Mortality in the
United States: Erratum.” Review of Economic Studies 73 (3): 847–847.
doi:http://restud.oxfordjournals.org/content/by/year.