Lecture
Forecasting & Long Memory: Time Series
In course
DEMO: voluptate minim pariatur irure
Et reprehenderit deserunt esse incididunt qui mollit labore et cillum occaecat exercitation nostrud nulla sint. Excepteur enim occaecat nostrud cillum id adipisicing aliquip labore proident et. Aute exercitation laboris cupidatat fugiat.
Description
This lecture covers autoregressive and moving average processes for forecasting in time series analysis. It delves into the implications of MA processes, invertibility, and forecasting for ARMA processes. Additionally, it explores the properties of forecast errors, predictive intervals, and error metrics. The second part of the lecture discusses long memory in stationary processes, the decay of autocovariances, and the concept of long-memory processes with polynomial decay.
Instructor
consequat dolore laborum anim
Cupidatat ea commodo incididunt nostrud cillum. Enim quis eu nostrud et deserunt dolore Lorem id sit mollit do in officia. Quis est eiusmod velit labore ut culpa eu esse ex.
Official source
About this result
This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.
Ontological neighbourhood
Related lectures (54)
Time Series Forecasting: ARMA Models
Covers ARMA models for time series forecasting, discussing implications, properties of forecast error, challenges with predictions, and covariance models.
Residuals & Forecasting: MATH-342 Time Series
Covers residuals, diagnostics, overfitting, and forecasting methods in time series analysis.
Model Choice and Prediction
Explores model choice, prediction, and forecasting techniques in time series analysis.
Time Series
Explores Time Series, covering model specification, diagnostics, and forecasting methods.
Model Selection in Time Series Analysis
Covers model selection, diagnostics, and forecasting in time series analysis, emphasizing the challenges of determining the model order based on autocorrelation and partial autocorrelation functions.