ploty_slideshow_example

---
title: "Plotly Slideshow Example"
author: "Miles McBain"
date: "6 August 2016"
output:   
  ioslides_presentation
---
```{r, include=FALSE}
library(AER)
library(ggplot2)
library(dplyr)
library(plotly)
```

#Welcome 

Example Deck

##Count data | Fixed Exposure
* **Exposure** refers to length of the observation interval in time/space/clicks/views/customers etc..
* Fixed exposure means it is the same for each observation
* Example data: [Number of tropical cyclones in Australian Region by year](http://www.bom.gov.au/cyclone/climatology/trends.shtml)

$$Y_{i} \sim Pois(\lambda_{i})$$ 
  $$\mu_{i} = \lambda_{i}$$
  $$g(\mu_{i}) = \eta_i$$
  
  for $g(.) = log(.)$ (canonical): $\mu_{i} = \exp(\eta_i)$  

##Count data | Variable Exposure
  * Variable exposure means we have some observation interval measure associated with each $y_{i}$
  + $n_{i}$, $t_{i}$... etc
* Example data: `AER::ShipAccidents` (next slide)

##Shipping Accidents

```{r, echo=FALSE}
data("ShipAccidents")
p <- ShipAccidents %>% 
  ggplot(aes(x=service, y=incidents, construction=construction)) + 
  xlab("service (months)") + 
  geom_point(aes(colour=type))
ggplotly(p)
```