Computer Science
8 years ago

New method accurately predicts future trends from time-series data.

1 view
IEEE Transactions on Automatic Control
Identification of autoregressive moving-average parameters of time series
Daniel Graupe, D. J. KRAUSE, Jason H. Moore

Paper Summary

The article presents a method to figure out the parameters and orders of mixed autoregressive moving-average signal models from time-series data. First, it identifies a purely autoregressive signal model. Then, it calculates the parameters and orders of the mixed autoregressive moving-average process using simple algebraic equations based on the purely autoregressive model parameters.

Related papers

Theme background image
Computer Science
a year ago

Deep learning outperforms traditional methods in time series analysis accuracy.

Time Series Clustering
Stock Market Predictions
Neural Networks
3 views
Theme background image
Computer Science
2 years ago

Non-autoregressive neural networks outperform autoregressive for faster system identification.

Neural Networks
Anomaly Detection
Image Denoising
3 views
Economics, Econometrics and Finance
4 years ago

New model accurately predicts unemployment trends, revolutionizing economic forecasting.

Macroeconomic Analysis
Regularization Methods
3 views