Auto Regressive Integrated Moving Average (ARIMA)
Time Series Forecasting
A time series is a sequence where a data is recorded over regular time intervals. It can be measured yearly, monthly, daily, or even hourly. ARIMA models and Exponential smoothing are the two most widely used approaches to time series forecasting. ARIMA models based on the autocorrelations in the data, while exponential smoothing models aim to describe the trend and seasonality in the data. Stationarity and differencing are two crucial parts of the time series. A stationary time series is one whose properties don’t depend on the time at which the series is observed. Computing the differences between consecutive observations in non-stationary data to make data stationery is known as differencing.
Example:
The dataset has been collected from the Kaggle. Kaggle is the world’s largest data science community. One can get the required data from here.
I have used the dataset of Food Pricing Monitoring to forecast the price based on historical records. Here is the link to the dataset.
https://www.kaggle.com/shirishsonvane/food-pricing-monitoring-tool-datasetUsing read.csv() function you can load the data into R studio.
ts <- read.csv("D:/PG/Stats/CA1/timeseries_final_data.csv")