New Introduction to Multiple Time Series Analysis

By Helmut Ltkepohl. 2007

6,260 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Management Science and Operations Research, Statistics and Probability, Time-Series & Forecasting

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Management Science and Operations Research

In Statistics and Probability

In Time-Series & Forecasting

Citations in the corpus

Causal Models for Dynamical Systems (2020, background): “, 2018b], mostly considering discrete time models such as vector autoregressive models [Lütkepohl, 2007], for example.”
Invariant Causal Prediction for Sequential Data (2017, methodology): “he time series setting, where the concept of Granger causality (Granger, 1969) plays a prominent role; see also structural vector auto-regressive models (SVAR) that are popular in econometrics (e.g., Lütkepohl, 2005). When excluding instantaneous eﬀects, the time order allows applying regression tec”
The Arrow of Time in Multivariate Time Series (2016, methodology): “For practical purposes, we restrict ourselves to VAR(p) processes; see Lütkepohl (2010) for the technical difficulties with fitting VAR(p, q) processes.”
High-Dimensional Multivariate Forecasting with Low-Rank Gaussian Copula Processes (2019, methodology): “The dataset obtained from [19] is preprocessed with the following steps similarly to [16]: • Data cleaning: removal of outliers in terms of average speed ( > 45 .”
Think Globally, Act Locally: A Deep Neural Network Approach to High-Dimensional Time Series Forecasting (2019, methodology): “For a comprehensive treatment of traditional methods, we refer the readers to [13, 19, 4, 18, 12] and the references there in.”
The Conditional Autoregressive Wishart Model for Multivariate Stock Market Volatility (2010, positive): “Following this idea, we decompose the scale matrix St for the daily covariance matrix Rt in Equations (1)-(3) into a secular component Mt and a mean-reverting short-run component S ∗ t : E(Rt|Ft−1) = St = CtS tC ′ t, with Mt = CtC ′ t, (24)”
Structural Vector Autoregressions (2011, methodology): “…changes in future productivity. subject to the first element of being zero and 0 = 1 where Ω() denotes the share of the forecast error variance of variable attributable to structural shock at horizon expressed in terms of the structural parameters of the model (also see Lütkepohl 2005).”
Path Forecast Evaluation (2010, background): “Other relevant references for speci c results on H include, among others, Clements and Hendry (1993) and Lütkepohl (2005).”
Regularized estimation in sparse high-dimensional time series models (2013, methodology): “Consistent estimation of VAR models in the low-dimensional regime ( p, d ﬁnite, T → ∞ ) relies on two assumptions (cf. Section 3.2.2, L¨utkepohl (2005)): (a) X (cid:48) X /N converges in probability to a non-singular matrix C (i.e., Λ min ( C ) > 0 ) and (b) vec ( X (cid:48) E ) /N converges to…”