futurera | open data, time series, forecasting

Introduction and Overview
Introduction and overview of forecasting methods
Time Series Regression
A time series regression forecasts a time series as a linear relationship with the independent variables.
Smoothing Methods
Smoothing methods are weighted averages of past observations.
- Simple moving averages
  assume that the future will equal the averages of the past
- Weigthed moving averages
  use different weights on each observation
- Single exponential smooting
  exponentially weighted moving averages
- Adaptive response-rate exponential smoothing
  exponentially weighted moving averages with automatically choosing the alpha
- Differences
  Estimating trends with differences
- Double Moving Averages
  Double Moving Averages
- Brown's Double Exponential Smoothing
  Brown's Double Exponential Smoothing
- Holt
  Holt's's Two Parameter Trend Model
- Winters
  Winters' Three-Parameter Exponential smoothing
Decomposition
Identify components, such as seasonality, trend, and cycle.
- Additive
  Time series data is a function of the sum of its components.
- Multiplicative
  As the data increase, so does the seasonal pattern.
Statistics
Metrics in forecasting
- Characteristic
  Characteristics of time series
- Error
  Error metrics in forecasting
- Performance
  Performance metrics in forecasting
- Miscellaneous
  Miscellaneous metrics in forecasting