BIN models and activities on financial markets
Faculty of Economics and Law (FSEJ),University Abdou Moumouni, Niamey BP 12442 Niamey – Niger.
Author’s E-mail: hassanemamoudou(at)hotmail.com
ACD: Autoregressive Conditional Durations; ARCH: Autoregressive Conditional Heteroscedasticity; ARMA: Autoregressive Moving Average; DGP: Data Generating Process; GARCH: General ARCH; MLE: Maximum Likelihood Estimation
In this paper we test empirically counting models for high frequency data: BIN(1,1) model with Poisson process, to verify the ability of the model to capture the clustering phenomenon in the case of high frequency data as regards stocks intraday data. The objectives of this paper are two-fold; on one hand we study the theoretical conceptualization of the BIN models, and on the other, tests are done to verify the adequacy of these models in comparison to the ARCH-GARCH model. The process of model estimation uses the data generating process (DGP) and the actual data of three stocks (Boeing, Disney and AWK) on the New York Stock Exchange. In this paper we study the issue of adequacy of BIN models to capture the activities of financial markets, about stocks intraday data (volume, quote, prices) and therefore aid in forecasting the evolution of financial markets activities. In fact, this model could be used in the Africa financial markets by contextualization.
Key words: ACD model, count data, BIN modesls, Poisson process, clustering, density forecast.
JEL classification: G14, G15.