Discover published sets by community

A model that uses the most recent data point as the next period's forecast. Typically used as a benchmark due to its simplicity.

Averages a number of the most recent actual data points to smooth out short-term fluctuations and highlight longer-term trends or cycles.

Similar to the moving average model but assigns weights to each data point based on its relevance, with more recent data typically given more weight.

A time series forecasting method that applies decreasing weights to past data exponentially. It’s useful for data with no clear trend or seasonal pattern.

An extension of exponential smoothing to capture trends by including a second equation that accounts for the trend component of the data.

Also known as Holt-Winters method, it includes a third equation to handle seasonal variations in addition to level and trend.

Stands for AutoRegressive Integrated Moving Average. ARIMA models time series data based on past values and errors; good for stationary datasets.

An extension of the ARIMA model that specifically takes into account seasonality in a time series dataset.

A forecasting technique for intermittent demand, where demand is infrequent and highly variable. It's based on separate exponential smoothing of demand levels and demand intervals.

Utilize statistical techniques to model economic data points and include variables like price, economic indicators, and marketing activities.

A statistical method for estimating the relationships among variables. Often used to forecast demand based on observed relationships between sales and other factors.

Uses qualitative data and expert opinion to make predictions when quantitative data is unavailable or to adjust quantitative forecasts.

A consensus technique where sales estimates from individual salespersons are reviewed for feasibility and combined to create a company-wide forecast.

Breaks down a time series into seasonal, trend, and random components. This model is good when a time series is influenced by seasonal factors.

A structured communication technique that relies on a panel of experts. The experts answer questionnaires in two or more rounds and the results are aggregated for the forecast.

Involves gathering data about consumers' needs and preferences to forecast potential demand. Often used for new products and services.

These models assume that the forecasted variable is influenced by one or more independent variables. Common techniques include regression and econometric models.

A form of machine learning that can model complex non-linear relationships. These are used for demand forecasting in cases where relationships between data points are too intricate for traditional models.

A statistical model that incorporates prior knowledge or beliefs alongside the observed data. It updates the forecasts as more data becomes available.