# Time Serie - Forecasting (Prediction)

### Table of Contents

## 1 - About

Forecasting is the process of making predictions of the future based on past and present data (time serie) and analysis of trends. Prediction is a similar, but more general term.

Usage can differ between areas of application: for example, in hydrology, the terms “forecast” and “forecasting” are sometimes reserved for estimates of values at certain specific future times, while the term “prediction” is used for more general estimates, such as the number of times floods will occur over a long period.

## 2 - Articles Related

## 3 - Methods

See Forecasting#Time_series_methods

- Autoregressive moving average (ARMA)
- Autoregressive integrated moving average (ARIMA) e.g. Box-Jenkins
- Extrapolation
- Growth curve (statistics)

See also: Time Serie - Model

### 3.1 - Trend Estimation (Regression)

Classical time-series prediction involves predicting the next n successive observations from a history of past observations. See Machine Learning - Linear (Regression|Model)

These problems have been studied extensively within the field of statistics (Brockwell & Davis 1996), but statistical techniques are only applicable when the data is limited to numerical features.

### 3.2 - Neural Network

Neural networks using a “sliding window” technique or a recurrent network architecture have also been successfully applied to time-series prediction problems, but also require numerical features to be effective (Biggus 1996).