A Calibration Procedure for Analyzing Stock Price Dynamics in an Agent-Based Framework

Abstract

In this paper we introduce a calibration procedure for validating of agent based models. Starting from the well-known financial model of Brock and Hommes 1998, we show how an appropriate calibration enables the model to describe price time series. We formulate the calibration problem as a nonlinear constrained optimization that can be solved numerically via a gradient-based method. The calibration results show that the simplest version of the Brock and Hommes model, with two trader types, fundamentalists and trend-followers, replicates nicely the price series of four different markets indices: the S&P 500, the Euro Stoxx 50, the Nikkei 225 and the CSI 300. We show how the parameter values of the calibrated model are important in interpreting the trader behavior in the different markets investigated. These parameters are then used for price forecasting. To further improve the forecasting, we modify our calibration approach by increasing the trader information set. Finally, we show how this new approach improves the model's ability to predict market prices. JEL codes: C53; C63; G17.

Agent-Based Model Calibration Using Machine Learning Surrogates

“…We find that the most relevant parameters to fit the empirical distribution of returns observed in the SP500 are those characterizing traders' attitude towards the trend (g 1 and g 2 ) and, secondly, their bias (b 1 and b 2 ). This result is in line with recent findings by Recchioni et al (2015) and Lamperti (2016) obtained using the same model. Moreover, the intensity of choice parameter (β, cf.…”

“…6 Finally, Recchioni et al (2015) use a simple gradient-based calibration procedure and then test the performance of the model they obtained through out of sample forecasting.…”

“…Despite the model is relatively simple, different contributions have tried to match the statistical properties of its output with those observed in real financial markets (Boswijk et al, 2007;Recchioni et al, 2015;Lamperti, 2016;Kukacka and Barunik, 2016). This makes the model an ideal test case for our surrogate: it is relatively cheap in terms of computational needs, it offers a reasonably large parameter space and it has been extensively studied in the literature.…”

Macroeconomic Policy in DSGE and Agent-Based Models Redux: New Developments and Challenges Ahead

“…Note that this latter technique has the merit of being potentially applicable to both the long-run equilibria of the model and during adjustment phases. Conversely, Recchioni et al (2015) approach the problem of calibrating the free parameters of ABMs as a nonlinear constrained optimization, which can be solved numerically via gradient-based methods, whereas Fabretti (2012) employs search technologies coming from genetic algorithms to explore the space of all possible parameter combinations in simple ABMs of financial markets. More recently, have suggested a Bayesian inference approach, as opposed to simulated minimum distance, to estimate ABM parameters, whereas Lamperti (2015Lamperti ( , 2016 resorts to information-criteria techniques to quantify the distance between the true probabilistic dynamics of the output of the model and the data (to be minimized in order to achieve estimation), without needing to impose any stationarity requirements.…”

Selection of the distributional rule as an alternative tool to foster cooperation in a Public Good Game

“…Janssen and Ostrom (2006) identify three useful environments in order to calibrate the model parameters: a) empirical data (see Recchioni et al (2015) and Alfarano et al (2007)), b) survey and/or interviews (Garcia et al (2007)), c) experiments (Roth and Erev (1995)). Moreover, different validation techniques have been also proposed (see Sargent (2013)).…”