Agent-based Model with Netlogo

Abstract

This report shows agent-based simulation models of car speeds that generate artificial time series data in comparison with real-world data. The model uses constructions of autoregression models, which are projected from both real-world and synthetic data with the help of a search algorithm. This report also validates procedures used in tackling the delinquent of confronting speculative replication models with data as well as the issue of comparing various models.

Introduction

The work shows the procedure to simulate data in by beginning with approximating the casual assembly build in the archetypal through the application of artificially produced data by the causal structure and the model fundamental to the real-world data. Additionally, it also compares with the inferred causal structure. On top of that, it provides a solution to the subjects of antagonizing an ABM to experimental data. It is possible to have a right corresponding between the underlying construction merged in the ABM and the development incorporated in the real-world data that offers particular practical provision. The fast upcoming of this move toward is dedicated to the key of such interconnected problems that can be observed as an indication of their sign and significance of the agent-based model.

The experiential counter-part is represented by a Structural Vector Autoregressive (SVAR) model (Alfarano, 2006). However, the RBC /DSGE approach is not relieved from challenging matter; structurally, firmness is based on personal conduct, rather than assuming a representative mediator which disregards or even rejects any inter-action forms. Additionally, the recognition of the empirical structure in the SVAR classical is usually accomplished by striking limitations resultant from the theoretic models, which are thus not subjected to any severe trial. (“Empirical validation of stochastic models of interacting agents.” In: The European Physical JournBarde, 2015)

The validation technique

This section explains simulation steps that comprise of five various procedures.

First, the simulated and empirical data are selected to be directly compared in the second phase of analyzing the developing properties of the series formed by the agent-based model.

This is the html code that was generated by netlogo

globals [

patel yogash

]

turtles-own [

speed

speed-limit

speed-min

]

to setup

clear-all

ask patches [ setup-road ]

setup-cars

watch sample-car

reset-ticks

end

to setup-road ;; patch procedure

if pycor < 2 and pycor > -2 [ set pcolor white ]

end

to setup-cars

if number-of-cars > world-width [

user-message (word

“There are too many cars for the amount of road. ”

“Please decrease the NUMBER-OF-CARS slider to below ”

(world-width + 1) ” and press the SETUP button again. ”

“The setup has stopped.”)

stop

]

set-default-shape turtles “car”

create-turtles number-of-cars [

set color blue

set xcor random-xcor

set heading 90

;; set initial speed to be in range 0.1 to 1.0

set speed 0.1 + random-float 0.9

set speed-limit 1

set speed-min 0

separate-cars

]

set sample-car one-of turtles

ask sample-car [ set color red ]

end

; this procedure is needed so when we click “Setup” we

; don’t end up with any two cars on the same patch

to separate-cars ;; turtle procedure

if any? other turtles-here [

fd 1

separate-cars

]

end

to go

;; if there is a car right ahead of you, match its speed then slow down

ask turtles [

let car-ahead one-of turtles-on patch-ahead 1

ifelse car-ahead != nobody

[ slow-down-car car-ahead ]

[ speed-up-car ] ;; otherwise, speed up

;; don’t slow down below speed minimum or speed up beyond speed limit

if speed < speed-min [ set speed speed-min ]

if speed > speed-limit [ set speed speed-limit ]

fd speed

]

tick

end

to slow-down-car [ car-ahead ] ;; turtle procedure

;; slow down so you are driving more slowly than the car ahead of you

set speed [ speed ] of car-ahead – deceleration

end

to speed-up-car ;; turtle procedure

set speed speed + acceleration

end

; Copyright 1997 Uri Wilensky.

; See Info tab for full copyright and license.

(wilensky, 2020)

Validation assessment

The final phase involves the comparison of the causal effects necessitated by the SVARRW and the SVARAB models. This process will elaborate on the number of  the real-world predicted causal effects that are seized under validation inquiry by the agent-based model. To compare the causal effects, the constructed similarity measure Ω, is applied starting from the approximations of the SVAR.

Conclusions

This paper validates the policy for oriented agent-based models. From those models, it is possible to create the artificial time series in comparison with the aggregates time series calculated by the netlogo. This action is grounded on comparing Structural Vector Autoregressive modes projected from real-world data as well as artificial data through causal search algorithms. Hence, the results should be compared with other agent-based models.

References

Alfarano. (2006). “Estimation of a sim-ple agent-based model of financial markets. newyork.

“Empirical validationof stochastic models of interacting agents”. In:The European Physical JournBarde, S. (2015). “Empirical validationof stochastic models of interacting agents”. In:The European Physical Journal B-CondensedMatter and Complex Systems. newyork.

wilensky. (2020). Retrieved from netlogo: http://www.netlogo.com