Sign in or start a free trial to avail of this feature.
1. Introduction to Time Series Forecasting
In this lesson, we introduce the concept of a time series forecast. We will look at the differences between time series analysis and regression analysis.
To explore more Kubicle data literacy subjects, please refer to our full library.
Lesson Goal (00:11)
The goal of this lesson is to understand the concept of time series forecasting.
Understanding Time Series Data (00:18)
Time series data refers to any single variable that has been observed over a period, indexed in time order. Our goal when analyzing time series data is to create a forecast, which is a projection of future values for the time series variable, based on the historical data.
The accuracy of a forecast depends on how much historical data is available, and how completely we understand what influences the results. As a result, the degree of confidence of a forecast can be more important than the forecast itself.
Time series forecasting has some similarities with regression analysis. Time series forecasting uses past values of a single variable to predict future values of the same variable, while regression uses the values of many variables at a single point in time to predict the value of another variable at the same point in time.
One important assumption made in time series forecasting is that the factors influencing a time series variable will be constant over time. In other words, factors affecting a variable in the past will affect the variable in the same way in the future. All the models we cover in this course make this assumption.
In this course, we'll look at some of the most popular forecasting techniques and look at how they're best deployed.
Our goal in this lesson is to introduce the concept of Time Series Forecasting.
Time series data is any single variable that has been observed over a period, indexed in time order.
This could be the height of the tide, the Euro dollar exchange rate, or the number of hits on a website landing page.
The Alteryx Time Series tools analyze such periodic data with the goal of forecasting future values. The accuracy of these forecasts depends on a number of factors.
These include how much historical data is available and how completely we understand what influences the results.
For this reason, we're often less concerned with the forecast itself and more concerned with the degree of confidence in our forecast.
Even the best forecasts are unlikely to predict the future precisely, but they can give you an indication of what results are more likely. Time Series Forecasting takes historical data and uses a data model to predict future values.
This has some similarities to regression analysis.
While a regression seeks to find the line of best fit between multiple vectors, in its simplest form, a time series forecast only requires a single set of data together with the corresponding dates or times.
For example, let's consider an online marketing data set.
We can see that this data set contains information on daily sales over a four-year period.
A regression analysis will use marketing spend by channel on a particular day to predict that day's sales. In contrast, a time series forecast looks at the sales each day and attempts to forecast future sales.
There is, of course, a degree of crossover between the two concepts.
When forecasting future sales, the model assumes that the factors influencing past sales will be consistent in the future. If marketing spend changes significantly for a certain channel, this will likely alter future sales.
Sophisticated models can account for these types of changes. However, we'll concentrate on more straightforward techniques.
We'll stop here for now. In the next lesson, we'll look at the different forecast methodologies available in Alteryx.