Glosarium AI
Kamus lengkap Kecerdasan Buatan
ARIMA/SARIMA Models
Autoregressive integrated moving average models for univariate time series forecasting with or without seasonal components.
Exponential Smoothing and ETS
Exponentially decreasing weighting techniques to model trend, seasonality, and errors in time series.
Recurrent Neural Networks
Deep learning architecture specialized in processing sequential data with temporal memory (RNN, LSTM, GRU).
Temporal Decomposition
Methods for separating a time series into trend, seasonality, and residual components (STL, X11, SEATS).
GARCH Volatility Models
Models for capturing and forecasting conditional heteroscedasticity and time-varying variance in financial series.
Multivariate Time Series
Analysis and forecasting of series with multiple interdependent variables using VAR, VECM, and state-space models.
Transformers for Time Series
Application of attention mechanisms and Transformer architectures for modeling long-term temporal dependencies.
Probabilistic Forecasting
Generation of complete forecast distributions rather than single points, including intervals and quantiles.
Temporal Anomaly Detection
Automatic identification of unusual patterns or outliers in continuous time series.
High-Frequency Time Series
Specialized techniques for analyzing and modeling data collected at very short intervals (milliseconds/seconds).
Bayesian Time Series Methods
Bayesian approaches for inference and forecasting with quantified uncertainty in time series.
Prophet and Additive Models
Additive decomposition models with automatic hyperparameter tuning for business-friendly forecasting.