parameterized-rmd-demo.Rmd

---
output:
  html_document:
    code_folding: hide
    toc: yes
    toc_float: yes
params:
  symbol: GOOG
  days:
    label: "Number of Previous Days"
    value: 90
    input: slider
    min: 30
    max: 360
  model:
    label: "Type of Forecasting Model"
    value: "First Difference - Arima 0,1,0"
    input: select
    choices: ["First Difference - Arima 0,1,0", "High Partial Autocorrelation - Arima 1,1,0", "Simple Exponential Smoothimg - Arima 0,1,1" ]
  forecast:
    label: "Desired Forecast Date"
    value: !r library(lubridate); Sys.Date() + days(1)
    input: date
---
  
# Report for `r params$symbol` on `r Sys.Date()`

## Summary{.tabset}

```{r, echo=TRUE, message=FALSE}
library(stringr)
library(lubridate)
library(quantmod)
library(DT)
library(dygraphs)
library(forecast)
library(highcharter)

prices <- round(getSymbols(params$symbol, auto.assign = FALSE), 2)
close <- Cl(last(prices))
open <- Op(last(prices))

recent <- last(prices, n=params$days)
recent_nv <- recent[,-5]

```

The stock closed `r ifelse(close>open,'up','down')` at `r close` dollars per share yesterday.

### Price History

The chart below is made with the `quantmod` and `highcharter` R packages. An API returns all of the price history based on the stock tick symbol provided as a parameter. The candlestick chart is a default function from highcharter, as is the the Economist theme.


```{r echo=FALSE}

highchart() %>% 
   hc_yAxis_multiples(
    list(title = list(text = NULL), height = "75%", top = "0%"),
    list(title = list(text = NULL), height = "15%", top = "80.5%", opposite = TRUE)
  ) %>%
  hc_add_series_ohlc(prices, yAxis=0, name= params$symbol) %>% 
  hc_add_series_xts(prices[,paste0(params$symbol,".Volume")], name="Volume", type="column", yAxis=1) %>% 
  hc_add_theme(hc_theme_economist())
```

### Raw Data

The table below displays the daily price data for `r params$symbol` for the last `r params$days` days. A concise, interactive table is created with the `DT` package. 

  
```{r echo=TRUE}
df <- as.data.frame(recent)
df[,paste0(params$symbol, ".Volume")] <- df[,paste0(params$symbol, ".Volume")]/1000000 
datatable(df) %>% 
  formatCurrency(c(paste0(params$symbol, ".Open"), paste0(params$symbol, ".High"), paste0(params$symbol, ".Low"), paste0(params$symbol,".Close"),
                  paste0(params$symbol, ".Adjusted")), digits=2) %>% 
  formatRound(c(paste0(params$symbol, ".Volume")), digits=0)
```

## Model

```{r, warning=FALSE, message=FALSE}
arima_coefs <- eval(parse(text = paste0("c(", str_extract(params$model, "\\d,\\d,\\d"), ")")))
m <- arima(recent[,1], arima_coefs)
f <- forecast(m, as.numeric(params$forecast - Sys.Date()))
```

The forecast for `r params$forecast` is `r round(as.numeric(f$mean)[1],2)` dollars.

This model is fit with the arima function in the forecast package. A `r params$model` model is used though in practice any range of models could be used.

```{r, warning=FALSE, message=FALSE}
plot(forecast(m,7), main="")
```