---
title: 'Module 7: lab'
author: 'BIOSTAT 213: Introduction to Computing in the R Software Environment'
date: 'Wednesday, February 27, 2019'
output: pdf_document
urlcolor: blue
---

Write your solutions in the provided code chunks. You may work in groups of 2--3.

1. Use the space below to import data and load libraries.

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
library(knitr)
#
```

1. *Use `nhanes.csv`.* Show the first five rows of low-density lipoprotein (LDL) cholesterol (`lbdhdd`), age (`ridageyr`,) gender (`gender`), history of smoking (`hs`), history of military service (`military`), and vigorous recreational activity (`vra`). Use `kable()` to output this data. Center-align all columns.

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
#
``` 

1. *Use `nhanes.csv`.* Use ggplot2 to create a scatter plot with LDL cholesterol (`lbdhdd`) on the $y$ axis and age (`ridageyr`) on the $x$ axis. Include a LOESS smoother.

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    ',out.width='50%'}
#
``` 

1. *Use `nhanes.csv`.* Fit a linear model, with LDL cholesterol (`lbdhdd`) as the outcome. Include adjustment for age(`age`,) gender (`gender`), history of smoking (`hs`), history of military service (`military`), and vigorous recreational activity (`vra`).

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
#
``` 

1. *Use `nhanes.csv`.* Use `kable()` to display the coefficient estimates, the standard errors, the test statistics, and the $p$-values. Omit the row representing the intercept. Center-align all columns. Round the first 3 columns to 1 digit and the last column to 2 digits. 

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
#
``` 

1. *Use `nhanes.csv`.* Use `cbind()` to create an object that contains the coefficient estimates and the lower and upper bounds of the 95% confidence intervals. Check the object type. Is this an object type we have discussed in this course? Use `as.data.frame()` to convert the object to a data frame. Use `names()` to give the columns new names: `Estimate`, `Lower`, and `Upper`. Omit the row representing the intercept. Use `kable()` to display the data frame. Center-align all columns and round all columns to 1 digit. 

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
#
``` 

1. *Use `nhanes.csv`.* Use the data frame defined immediately above to check whether each confidence interval includes 0.

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
#
``` 

1. *Use `nhanes.csv`.* In Module 6's lab, you were asked to use `tapply()` and a custom function to find the percent of individuals in each `bpcat` group who smoke (`hs`). What is a more efficient way to find the percent of individuals in each `bpcat` group who smoke (`hs`)? Show this method. 

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
#
``` 

1. *Use `nhanes.csv`.* Write a function that takes as an argument a vector (you can think of the vector as a column in a data frame). The function should return the number of missing values that the variable contains. Check the function on the `ridageyr` and `bmxwt` variables.

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
#
``` 

1. *Use `nhanes.csv`.* Use one of the apply functions to apply the above function to every single column in the NHANES data. 

```{r,prompt=TRUE,comment=NA,collapse=TRUE,indent='    '}
#
```