analysis-3.Rmd

---
runtime: shiny
output: html_document
---
# Analysis 3: RMarkdown notebook rendered in RShiny with interactive widgets

## Funky interactive widgets, maaaaaan! 🤘
```{r echo = FALSE}
selectInput("n_breaks", label = "Number of wheelie bins:",
              choices = c(10, 20, 35, 50), selected = 20)

sliderInput("bw_adjust", label = "Bandwidth ... width:",
              min = 0.2, max = 2, value = 1, step = 0.2)
```

```{r echo=FALSE, message=FALSE}
library(tidyverse)
library(ggplot2)
library(patchwork)
```

## Here is the same analysis done in the other notebooks

Load the data in `data/`:
```{r echo=TRUE}
indir = 'data/'

fname_1 = 'Sep2021_Liaoning_data.csv'
fname_2 = 'Sep2021_Jilin_data.csv'
fname_3 = 'Sep2021_Heilongjiang_data.csv'

provs = c("Liaoning", "Jilin", "Heilongjiang")

yield_1 = read.csv(paste(indir,fname_1, sep=""), header = TRUE,sep = ",", stringsAsFactors = FALSE)
yield_2 = read.csv(paste(indir,fname_2, sep=""), header = TRUE,sep = ",", stringsAsFactors = FALSE)
yield_3 = read.csv(paste(indir,fname_3, sep=""), header = TRUE,sep = ",", stringsAsFactors = FALSE)
```

Create a function that creates the plots:
```{r echo=TRUE}
scatter_trend_plot = function(data, x, y, title, xlab="var", ylab="Normalised maize yield", method=lm) {
  plot = ggplot(data=data, aes(x={{x}}, y={{y}})) +
    geom_point() +
    geom_smooth(method={{method}}) +
    labs(title=title, x=xlab, y=ylab)
  return(plot)
}
```

Generate the plots for each of the datasets:
```{r echo=TRUE}
ljun = scatter_trend_plot(data=yield_1, 
                          x=Jun.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[1], 
                          xlab="June Temperature / degrees C",)
ljul = scatter_trend_plot(data=yield_1, 
                          x=Jul.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[1], 
                          xlab="July Temperature / degrees C",)
laug = scatter_trend_plot(data=yield_1, 
                          x=Aug.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[1], 
                          xlab="August Temperature / degrees C",)
jjun = scatter_trend_plot(data=yield_2, 
                          x=Jun.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[2], 
                          xlab="June Temperature / degrees C",)
jjul = scatter_trend_plot(data=yield_2, 
                          x=Jul.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[2], 
                          xlab="July Temperature / degrees C",)
jaug = scatter_trend_plot(data=yield_2, 
                          x=Aug.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[2], 
                          xlab="August Temperature / degrees C",)
hjun = scatter_trend_plot(data=yield_3, 
                          x=Jun.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[3], 
                          xlab="June Temperature / degrees C",)
hjul = scatter_trend_plot(data=yield_3, 
                          x=Jul.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[3], 
                          xlab="July Temperature / degrees C",)
haug = scatter_trend_plot(data=yield_3, 
                          x=Aug.T, 
                          y=Maize.yield.anomaly, 
                          title=provs[3], 
                          xlab="August Temperature / degrees C",)
```

Then assemble the plots together using `patchwork`
```{r echo=TRUE, message=FALSE, warning=FALSE}
(ljun+ljul+laug)/(jjun+jjul+jaug)/(hjun+hjul+haug)
```