Stat 552

Homework 6

Due Feb 26th

Reading: Faraway 6 & 7

  1. Faraway 6.1
  2. Faraway 7.4 & 7.5
  3. Is there evidence heart disease is related to pace of life?

    The dataset ex0914 in the package Sleuth3 contains observations on three pace of life indicators and heart disease in cities across America. A more thorough description is attached here: pace-of-life

    Your task is to answer the question “Is there evidence heart disease is related to pace of life?” through a complete regression analysis.

    Your answer should be a complete report with the following sections:

    • Introduction Give a brief overview of the data, a little bit of background and the questions of interest. Keep this concise, understandable to someone outside of this class, free of statistical jargon and to the point. You should provide a summary graphic of the data involved or some basic summary statistics (it’s up to you how you do this).

    • Methods Set down any regression models you plan to fit and describe how the model(s) will be used to answer the question of interest. State the assumptions required for inference.

    • Results Discuss your examinations of the diagnostics and unusual observations and and any adjustments that were made to the model or data. Provide your findings including:

      • fitted models - including estimate’s, s.e.’s, confidence intervals (if relevant), the estimate of , , adjusted and degrees of freedom.
      • the results of hypothesis tests (if relevant)
    • Conclusions Provide interpretation of the relevant results in the context of the study. Also set out model and/or design/data limitations and suggestions for future work.

    The report should be generated using Rmarkdown (or knitr or Sweave etc.) but should not include any raw R code or raw R output.

    Make sure graphs and figures are labelled properly (i.e. using English language names for axis labels not abbreviated R variable names).

    library(Sleuth3)
    ex0914$city <- c("Boston MA", "Buffalo NY", "New York NY", "Salt Lake City UT", 
      "Columbus OH", "Worcester MA", "Providence RI", "Springfield MA", 
      "Rochester NY", "Kansas City MO", "St. Louis MO", "Houston TX", 
      "Paterson NJ", "Bakersfield CA", "Atlanta GA", "Detroit MI", 
      "Youngstown OH", "Indianapolis IN", "Chicago IL", "Philadelphia PA", 
      "Louisville KY", "Canton OH", "Knoxville TN", "San Francisco CA", 
      "Chattanooga TN", "Dallas TX", "Oxnard CA", "Nashville TN", "San Diego CA", 
      "East Lansing MI", "Fresno CA", "Memphis TN", "San Jose CA", 
      "Shreveport LA", "Sacramento CA", "Los Angeles CA")
    head(ex0914)
    
    ##   Bank Walk Talk Heart              city
    ## 1   31   28   24    24         Boston MA
    ## 2   30   23   23    29        Buffalo NY
    ## 3   29   24   18    31       New York NY
    ## 4   28   28   23    26 Salt Lake City UT
    ## 5   27   22   30    26       Columbus OH
    ## 6   26   25   24    20      Worcester MA
    
You might find the `pander` function in the `pander` package useful for nicely typesetting
tables in Rmarkdown.