## Example Code for Regression
The data are in the library `lasso2`.  You will need to install the package from CRAN (click on the Packages tab in Rstudio) and then load the library:
```{r library}
library(lasso2)
```

We can load the prostate data from the package
```{r data}
data(Prostate)    # n = 97, 9 variable
names(Prostate)
```
The `names` function provides the names of the variables in the dataframe. 
To get a sense of the data use the `summary` function
```{r summary}
summary(Prostate)
```

## Fitting a Linear Model

The `lm` function is used to fit linear models
```{r lm}
prostate.lm = lm(lpsa ~ ., data=Prostate)
```
where the first argument is a model formula.  Here `lpsa` is the log of PSA, a marker for Prostate cancer.  The `.` to the right of the `~` indicates that all variables are to be included in the model, while the `data=Prostate` provides the name of the dataframe for the data.

The summaries of the MLEs are obtained using the summary function
```{r sum}
summary(prostate.lm)
```

The object `prostate.lm` has many summaries stored that you may extract. To see what is available use
```{r obj}
names(prostate.lm)
```
To find the residual degrees of freedom, for example, use `prostate.lm$df.residual`

To obtain 95\% confidence intervals we may either use $\hat{\beta}_j \pm SE(\beta_j) t_{.025}$ 

```{r man}
prostate.coef = summary(prostate.lm)$coef
t = qt(.975, df=prostate.lm$df.residual)
cbind(prostate.coef[,1] - t*prostate.coef[,2], prostate.coef[,1] + t*prostate.coef[,2])
```


or with the `confint` function:
```{r ci}

confint(prostate.lm)
```

or nicely formatted 

```{r}
knitr::kable(confint(prostate.lm))
```

For predictions the `predict` function provides point estimates and SE.

```{r predict}

prostate.fit = predict(prostate.lm, se=T)
prostate.fit
```


Rather than using Rmarkdown, explore using knitr to create pdf documents with nice tables.   (the packages `xtable` or `stargazer` are helpful!)