Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
94 tokens/sec
Gemini 2.5 Pro Premium
55 tokens/sec
GPT-5 Medium
38 tokens/sec
GPT-5 High Premium
24 tokens/sec
GPT-4o
106 tokens/sec
DeepSeek R1 via Azure Premium
98 tokens/sec
GPT OSS 120B via Groq Premium
518 tokens/sec
Kimi K2 via Groq Premium
188 tokens/sec
2000 character limit reached

Margin-closed vector autoregressive time series models (2211.11898v2)

Published 21 Nov 2022 in stat.ME

Abstract: Conditions are obtained for a Gaussian vector autoregressive time series of order $k$, VAR($k$), to have univariate margins that are autoregressive of order $k$ or lower-dimensional margins that are also VAR($k$). This can lead to $d$-dimensional VAR($k$) models that are closed with respect to a given partition ${S_1,\ldots,S_n}$ of ${1,\ldots,d}$ by specifying marginal serial dependence and some cross-sectional dependence parameters. The special closure property allows one to fit the sub-processes of multivariate time series before assembling them by fitting the dependence structure between the sub-processes. We revisit the use of the Gaussian copula of the stationary joint distribution of observations in the VAR($k$) process with non-Gaussian univariate margins but under the constraint of closure under margins. This construction allows more flexibility in handling higher-dimensional time series and a multi-stage estimation procedure can be used. The proposed class of models is applied to a macro-economic data set and compared with the relevant benchmark models.

Citations (1)

Summary

We haven't generated a summary for this paper yet.

Dice Question Streamline Icon: https://streamlinehq.com

Follow-up Questions

We haven't generated follow-up questions for this paper yet.

Don't miss out on important new AI/ML research

See which papers are being discussed right now on X, Reddit, and more:

“Emergent Mind helps me see which AI papers have caught fire online.”

Philip

Philip

Creator, AI Explained on YouTube