Many of you have faced challenges when it comes to time series forecasting and have been requesting our mentors to help you with the topic. Hence, here we are trying to debug some of the most common doubts that you face.
What is time series forecasting?
Time series forecasting is an important area of business analytics. The importance of time series forecasting lies in the fact that a lot of prediction tasks involve a component of time.
‘A time series is a sequence of observations taken sequentially in time.’
Any dataset is actually a collection of observations with time having an important role to play.
This concept has been covered extensively in our Business Analytics Nanodegree program. In fact, time series forecasting is one of the most popular topics of discussion among our students.
One of the most frequently asked questions is from the video game demand forecast project.
Question – If you look at the 2nd graph in the decomposition plot (seasonal data is on the Y axis). It BARELY increases over the year. Why is this considered ‘seasonality changes in magnitude each year’?
Here’s an answer which might be useful for you – “The seasonal portion shows that the regularly occurring spike in sales each year changes in magnitude, even so slightly rather than being constant. In Alteryx, we will need to hover the mouse over the seasonal graph in Interface mode to be able to see that the seasonal numbers are slightly increasing. So we have change in magnitude.
This is important because:
– Having seasonality suggests that any ARIMA models used for analysis will need seasonal differencing
– The change in magnitude suggests that any ETS models will use a multiplicative method in the seasonal component.”
If any of the steps are still unclear, you might also find it useful to contact your mentor. Additionally, here you can access some of the previously asked questions.