Stable Multiple Time Step Simulation/Prediction from Lagged Dynamic Network Regression Models *

Abhirup Mallik, Zack W Almquist

Research output: Book/ReportCommissioned report

Abstract

Recent developments in computers and automated data collection strategies have greatly increased the interest in statistical modeling of dynamic networks. Many of the statistical models employed for inference on large-scale dynamic networks suffer from limited forward simulation/prediction ability. A major problem with many of the forward simulation procedures is the tendency for the model to become degenerate in only a few time steps, i.e., the simulation/prediction procedure results in either null graphs or complete graphs. Here, we describe an algorithm for simulating a sequence of networks generated from lagged dynamic network regression models DNR(V), a sub-family of TERGMs. We introduce a smoothed estimator for forward prediction based on smoothing of the change statistics obtained for a dynamic network regression model. We focus on the implementation of the algorithm, providing a series of motivating examples with comparisons to dynamic network models from the literature. We find that our algorithm significantly improves multi-step prediction/simulation over standard DNR(V) forecasting. Furthermore, we show that our method performs comparably to existing more complex dynamic network analysis frameworks (SAOM and STERGMs) for small networks over short time periods, and significantly outperforms these approaches over long time time intervals and/or large networks.
Original languageEnglish (US)
StatePublished - 2018

Fingerprint

Electric network analysis
Statistics
Statistical Models

Keywords

  • DNR
  • DNRV
  • Dynamic Networks
  • ERGM
  • logistic re-gression
  • logit
  • network simulation
  • TERGM

Cite this

Stable Multiple Time Step Simulation/Prediction from Lagged Dynamic Network Regression Models *. / Mallik, Abhirup; Almquist, Zack W.

2018.

Research output: Book/ReportCommissioned report

@book{02c6301cf13b4acabedfd95c89261ef6,
title = "Stable Multiple Time Step Simulation/Prediction from Lagged Dynamic Network Regression Models *",
abstract = "Recent developments in computers and automated data collection strategies have greatly increased the interest in statistical modeling of dynamic networks. Many of the statistical models employed for inference on large-scale dynamic networks suffer from limited forward simulation/prediction ability. A major problem with many of the forward simulation procedures is the tendency for the model to become degenerate in only a few time steps, i.e., the simulation/prediction procedure results in either null graphs or complete graphs. Here, we describe an algorithm for simulating a sequence of networks generated from lagged dynamic network regression models DNR(V), a sub-family of TERGMs. We introduce a smoothed estimator for forward prediction based on smoothing of the change statistics obtained for a dynamic network regression model. We focus on the implementation of the algorithm, providing a series of motivating examples with comparisons to dynamic network models from the literature. We find that our algorithm significantly improves multi-step prediction/simulation over standard DNR(V) forecasting. Furthermore, we show that our method performs comparably to existing more complex dynamic network analysis frameworks (SAOM and STERGMs) for small networks over short time periods, and significantly outperforms these approaches over long time time intervals and/or large networks.",
keywords = "DNR, DNRV, Dynamic Networks, ERGM, logistic re-gression, logit, network simulation, TERGM",
author = "Abhirup Mallik and Almquist, {Zack W}",
year = "2018",
language = "English (US)",

}

TY - BOOK

T1 - Stable Multiple Time Step Simulation/Prediction from Lagged Dynamic Network Regression Models *

AU - Mallik, Abhirup

AU - Almquist, Zack W

PY - 2018

Y1 - 2018

N2 - Recent developments in computers and automated data collection strategies have greatly increased the interest in statistical modeling of dynamic networks. Many of the statistical models employed for inference on large-scale dynamic networks suffer from limited forward simulation/prediction ability. A major problem with many of the forward simulation procedures is the tendency for the model to become degenerate in only a few time steps, i.e., the simulation/prediction procedure results in either null graphs or complete graphs. Here, we describe an algorithm for simulating a sequence of networks generated from lagged dynamic network regression models DNR(V), a sub-family of TERGMs. We introduce a smoothed estimator for forward prediction based on smoothing of the change statistics obtained for a dynamic network regression model. We focus on the implementation of the algorithm, providing a series of motivating examples with comparisons to dynamic network models from the literature. We find that our algorithm significantly improves multi-step prediction/simulation over standard DNR(V) forecasting. Furthermore, we show that our method performs comparably to existing more complex dynamic network analysis frameworks (SAOM and STERGMs) for small networks over short time periods, and significantly outperforms these approaches over long time time intervals and/or large networks.

AB - Recent developments in computers and automated data collection strategies have greatly increased the interest in statistical modeling of dynamic networks. Many of the statistical models employed for inference on large-scale dynamic networks suffer from limited forward simulation/prediction ability. A major problem with many of the forward simulation procedures is the tendency for the model to become degenerate in only a few time steps, i.e., the simulation/prediction procedure results in either null graphs or complete graphs. Here, we describe an algorithm for simulating a sequence of networks generated from lagged dynamic network regression models DNR(V), a sub-family of TERGMs. We introduce a smoothed estimator for forward prediction based on smoothing of the change statistics obtained for a dynamic network regression model. We focus on the implementation of the algorithm, providing a series of motivating examples with comparisons to dynamic network models from the literature. We find that our algorithm significantly improves multi-step prediction/simulation over standard DNR(V) forecasting. Furthermore, we show that our method performs comparably to existing more complex dynamic network analysis frameworks (SAOM and STERGMs) for small networks over short time periods, and significantly outperforms these approaches over long time time intervals and/or large networks.

KW - DNR

KW - DNRV

KW - Dynamic Networks

KW - ERGM

KW - logistic re-gression

KW - logit

KW - network simulation

KW - TERGM

M3 - Commissioned report

BT - Stable Multiple Time Step Simulation/Prediction from Lagged Dynamic Network Regression Models *

ER -