Published in · 3 min read · May 2, 2020
--
Wouldn’t it be wonderful to be able to see the future? We will probably get there someday, but time series forecasting gets you close as it allows us to look ahead of time and helps us in business forecasting.
Some forecasting methods are extremely simple and surprisingly effective. While there are a wide range of forecasting methods, in this article we focus on three simple methods that financial analysts use to predict future revenues, expenses, and capital costs for a business etc. They are: (1) Average, (2) Naïve, and (3) Seasonal Naïve.
(1) Average Method:
Here, the forecasts or the future predictions are equal to the average or mean of the historical data available.
We can forecast this using meanf function in R as shown below :
meanf(y, h)
y -> Contains the time series data
h -> Forecast Horizon
Example: The below plot shows the average or mean method applied to forecast the Australian quarterly beer production.
Note : R code for all the example plots in this article can be found here
(2) Naïve Method:
In Naïve method, we simply set all the forecasts to the latest observation values.
In general, a person or an action showing a lack of experience is termed as Naïve. So, this method would work very well for many economic and financial time series data where there is no related historical data.
We can forecast this using naive function in R as shown below :
naive(y, h)
Example: The below plot shows the Naive method applied to forecast the Australian quarterly beer production.
Note : R code for all the example plots in this article can be found here
(3) Seasonal Naïve Method:
