Forecasting Strategies

So far, we have mainly been focusing on the first two components of the time series analysis workflow—data preprocessing and descriptive analysis. Starting from this chapter, we will shift gear and move on to the third and last component of the analysis—the forecast. Before we dive into different forecasting models in the upcoming chapters, we will introduce the main elements of the forecasting workflow. This includes approaches for training a forecasting model, performance evaluation, and benchmark methods. This will provide you with a set of tools for designing and building a forecasting model according to the goal of the analysis.

This chapter covers the following topics:

• Training and testing approaches for a forecasting model
• Performance evaluation methods and error measurement matrices
• Benchmark methods
• Quantifying forecast uncertainty...

Technical requirement

The following packages will be used in this chapter:

• forecast: Version 8.5 and above
• TSstudio: Version 0.1.4 and above
• plotly: Version 4.8 and above

You can access the codes for this chapter from the following link:

https://github.com/PacktPublishing/Hands-On-Time-Series-Analysis-with-R/tree/master/Chapter08

The forecasting workflow

Traditional time series forecasting follows the same workflow as most of the fields of predictive analysis, such as regression or classification, and typically includes the following steps:

1. Data preparation: Here, we prepare the data for the training and testing process of the model. This step includes splitting the series into training (in-sample) and testing (out-sample) partitions, creating new features (when applicable), and applying a transformation if needed (for example, log transformation, scaling, and so on).
2. Train the model: Here, we used the training partition to train a statistical model. The main goal of this step is to utilize the training set to train, tune, and estimate the model coefficients that minimize the selected error criteria (later on in this chapter, we will discuss common error metrics in detail). The fitted values and the model...

Training approaches

One of the core elements of the forecasting workflow is the model training process. The quality of the model's training will have a direct impact on the forecast output. The main goals of this process are as follows:

• Formalize the relationship of the series with other factors, such as seasonal and trend patterns, correlation with past lags, and external variables in a predictive manner
• Tune the model parameters (when applicable)
• The model is scalable on new data, or in other words, avoids overfitting

As we mentioned previously, prior to the training process, the series is split into training and testing partitions, where the model is being trained on the training partition and tested on the testing partition. These partitions must be in chronological order, regardless of the training approach that has been used. The main reason for this is that most...

Finalizing the forecast

Now that the model has been trained, tested, tuned (if required), and evaluated successfully, we can move forward to the last step and finalize the forecast. This step is based on recalibrating the model's weights or coefficients with the full series. There are two approaches to using the model parameter setting:

• If the model was tuned manually, you should use the exact tuning parameters that were used on the trained model
• If the model was tuned automatically by an algorithm (such as the auto.arima function we used previously), you can do either of the following:
  • Extract the parameter setting that was used by with the training partition
  • Let the algorithm retune the model parameters using the full series, under the assumption that the algorithm has the ability to adjust the model parameters correctly when training the model with new data

The use...

Handling forecast uncertainty

The main goal of the forecasting process, as we saw previously, is to minimize the level of uncertainty around the future values of the series. Although we cannot completely eliminate this uncertainty, we can quantify it and provide some range around the point estimate of the forecast (which is nothing but the model's expected value of each point in the future). This can be done by using either the confidence interval (or a credible interval, when using the Bayesian model) or by using simulation.

Confidence interval

The confidence interval is a statistical approximation method that's used to express the range of possible values that contain the true value with some degree of confidence...

Summary

The training process of a forecasting model is the final step of the time series analysis. The focus of this chapter was to introduce the principle of the forecasting workflow. As we saw, there are several methods that we can use to train a forecasting model, and the method selection process should align with the forecasting goals and available resources. In the following chapters, you will see these applications in practice.

In the next chapter, we will use the applications of the linear regression model to forecast time series data.