A Comparison of Three Models to Predict Liquidity Flows between Banks Based on Daily Payments Transactions

Ron Triepels; Hennie Daniels

doi:10.2139/ssrn.2835979

A Comparison of Three Models to Predict Liquidity Flows between Banks Based on Daily Payments Transactions

Ron Triepels^*, Hennie Daniels

^*Corresponding author for this work

Research output: Working paper / Preprint › Discussion paper

Abstract

The analysis of payment data has become an important task for operators and overseers of financial market infrastructures. Payment data provide an accurate description of how banks manage their liquidity over time. In this paper we compare three models to predict future liquidity flows from payment data: 1) a moving average model, 2) a linear dynamic system that links the inflow of banks with their outflow, and 3) a similar dynamic system but with a constraint that guarantees the conservation of liquidity. The error graphs of one-step-ahead predictions on real-world payment data reveal that the moving average model performs best, followed by the dynamic system with constraint, and finally the dynamic system without constraint.

Original language	English
Publisher	CentER
Number of pages	16
DOIs	https://doi.org/10.2139/ssrn.2835979
Publication status	Published - 2016
Externally published	Yes

Publication series

Series	CentER Discussion Paper Series
Number	037

JEL classifications

c32 - "Multiple or Simultaneous Equation Models: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models"
c53 - "Forecasting and Prediction Methods; Simulation Methods "
c61 - "Optimization Techniques; Programming Models; Dynamic Analysis"
e42 - "Monetary Systems; Standards; Regimes; Government and the Monetary System; Payment Systems"
e44 - Financial Markets and the Macroeconomy
e47 - Money and Interest Rates: Forecasting and Simulation: Models and Applications

Keywords

large-value payment systems
predictive modeling
dynamic system
TIME-SERIES ANALYSIS

Access to Document

10.2139/ssrn.2835979

Cite this

@techreport{98f2f765154241789d97a9b570ca78a7,

title = "A Comparison of Three Models to Predict Liquidity Flows between Banks Based on Daily Payments Transactions",

abstract = "The analysis of payment data has become an important task for operators and overseers of financial market infrastructures. Payment data provide an accurate description of how banks manage their liquidity over time. In this paper we compare three models to predict future liquidity flows from payment data: 1) a moving average model, 2) a linear dynamic system that links the inflow of banks with their outflow, and 3) a similar dynamic system but with a constraint that guarantees the conservation of liquidity. The error graphs of one-step-ahead predictions on real-world payment data reveal that the moving average model performs best, followed by the dynamic system with constraint, and finally the dynamic system without constraint.",

keywords = "large-value payment systems, predictive modeling, dynamic system, TIME-SERIES ANALYSIS",

author = "Ron Triepels and Hennie Daniels",

year = "2016",

doi = "10.2139/ssrn.2835979",

language = "English",

series = "CentER Discussion Paper Series",

publisher = "CentER",

number = "037",

edition = "37",

type = "WorkingPaper",

institution = "CentER",

}

TY - UNPB

T1 - A Comparison of Three Models to Predict Liquidity Flows between Banks Based on Daily Payments Transactions

AU - Triepels, Ron

AU - Daniels, Hennie

PY - 2016

Y1 - 2016

N2 - The analysis of payment data has become an important task for operators and overseers of financial market infrastructures. Payment data provide an accurate description of how banks manage their liquidity over time. In this paper we compare three models to predict future liquidity flows from payment data: 1) a moving average model, 2) a linear dynamic system that links the inflow of banks with their outflow, and 3) a similar dynamic system but with a constraint that guarantees the conservation of liquidity. The error graphs of one-step-ahead predictions on real-world payment data reveal that the moving average model performs best, followed by the dynamic system with constraint, and finally the dynamic system without constraint.

AB - The analysis of payment data has become an important task for operators and overseers of financial market infrastructures. Payment data provide an accurate description of how banks manage their liquidity over time. In this paper we compare three models to predict future liquidity flows from payment data: 1) a moving average model, 2) a linear dynamic system that links the inflow of banks with their outflow, and 3) a similar dynamic system but with a constraint that guarantees the conservation of liquidity. The error graphs of one-step-ahead predictions on real-world payment data reveal that the moving average model performs best, followed by the dynamic system with constraint, and finally the dynamic system without constraint.

KW - large-value payment systems

KW - predictive modeling

KW - dynamic system

KW - TIME-SERIES ANALYSIS

U2 - 10.2139/ssrn.2835979

DO - 10.2139/ssrn.2835979

M3 - Discussion paper

T3 - CentER Discussion Paper Series

BT - A Comparison of Three Models to Predict Liquidity Flows between Banks Based on Daily Payments Transactions

PB - CentER

ER -

A Comparison of Three Models to Predict Liquidity Flows between Banks Based on Daily Payments Transactions

Abstract

Publication series

JEL classifications

Keywords

Access to Document

Fingerprint

Cite this