Reference no: EM133009458
Assignment: Question
This. dataset contains data from a survey of people aged over 66 who live in aged-care facilities in the U. The variables are:
• hospital = Number of hospital stays.
• health = self-perceived health status, levels are "poor", "average", "excellent".
• chronic = Number of chronic conditions.
• adi = whether the individual has a condition that limits activities of daily living ("limited") or not ("normal").
• region = levels are northeast, midwest, west, other.
• age = Age in years (divided by 10).
• afam = Is the individual African-American?
• gender = indicating gender.
• married = is the individual married?
• school = Number of years of education.
• income = Family income in USD 10000.
• employed = Is the individual employed?
• insurance = Is the individual covered by private insurance?
• medicaid = Is the individual covered by Medicaid?
Your task is to use methods taught in this unit to construct a model to predict the number of hospital stays. Some particular questions of interest are the following:
1. Some of the variables in the dataset provide direct information about the health of the individual. The others are demographic variables.
You must write and submit a brief report that explains and describes the construction of your model. it should contain all the information necessary for the marker to understand what you have clone and why. It should also provide answers to the above questions. Your report should be written in the style of a university essay, and not exceed 1500 words (the word count excludes R code, tables, plots, etc, but includes footnotes). It should not include appendices.
You should proofread your work and ensure that the spelling and grammar are correct.
Your report should be written using RMarkdown and rendered as an html file. It should contain all the R code used to generate the results. This page provides some tips on writing reports in RMarkdown. You must submit 2 documents:
1. An RMarkdown file which, when run, generates your html file. This will not be separately marked, but may be used by the marker to check your work.
2. The html file.
The RMarkdown file will not be marked separately, but may be used by the marker to examine your work in greater detail than is provided in your report.
Your task is to use methods taught in this unit to construct a model to predict the number of hospital stays. Some particular questions of interest are the following:
1. Some of the variables in the dataset provide direct information about the health of the individual. The others are demographic variables. If the 'health' variables are included in your model, are the demographic variables then much help in making predictions of hospital stays?
2. Does insurance coverage (either private insurance or medicaid) affect the number of hospital stays?
3. Consider a 75 year old who has poor health, 2 chronic conditions and medicaid. Predict the probability that this person will have at least one hospital stay. Also, predict the probability that this person will have more than one hospital stay. Note: any other explanatory variables in your model should be set to 'sensible' values when computing your prediction.
4. Consider a 66 year old who has excellent health, no chronic conditions and no health insurance. Predict the probability that this person will have no hospital stays. Also, predict the probability that this person will have exactly one hospital stay. Note: any other explanatory variables in your model should be set to 'sensible' values when computing your prediction.
You must write and submit a brief report that explains and describes the construction of your model. It should contain all the information necessary for the marker to understand what you have clone and why.
Attachment:- Demographic variable.rar