R Statistics Cookbook

How to do it...

We will use the anomalize package to detect outliers for a time series containing the amount of dollars sold to the market by the agricultural-export sector.

1. First, we load the necessary libraries:

library(anomalize) library(dplyr) library(tibbletime) 

2. We load the time series for US dollars sold by farmers in Argentina:

currency_sales = read.csv("./sold_usd_oilseeds.csv") currency_sales$indice_tiempo = as.Date(currency_sales$indice_tiempo,"%Y-%m-%d") 

3. We then do the anomaly detection, which has several components. First, we transform the dataset into a tibbletime object, and then we decompose the time series into trend, seasonality, and remainder components. Second, we detect the anomalies using the remainder, and we ...