compute_comp_lik: Computes the likelihood of the compressed EKOP model

In simonsays1980/bayespin: An R package for Bayesian estimation of the probability of informed trading

Description

Calling compute_comp_lik() computes the likelihood of the compressed EKOP model. The compressed EKOP model has been first proposed by Jackson (2007) and uses only the number of trades per trading day instead of the number of buyer- and seller-initiated trades per day.

Usage

compute_comp_lik(data, par, T, methodLik = c("precise", "approx"))

Arguments

data	A vector containing the trades data.
par	A vector specifying the parameter values at which the function should be evaluated. The parameter order is alpha, epsilon, delta, and mu.
T	A double indicating the length of a trading day in minutes.
methodLik	A character specifying, if undefined function values in optimization should be approximated by large defined values (1e+6). This can help to make maximum likelihood estimation more stable.

Details

In case of large trading volumes the function computation suffers sometimes from number overflow. In this case the parameter methodLik offers an approximation method in which Inf or NaN values are replaced by the value 1e+6 to avoid number overflow. In experiments this approximation has shown very good performance and we suggest the user to consider this approximation when function calculation fails to number overflow.

Value

A double holding the likelihood value of the data and parameter values passed in.

References

• Jackson, D., 2007. Infering trader behavior from transaction data: A trade count model. Journal of Computational and Graphical Statistics 12, 55-79.

See Also

• estimate_compml() for the calling function

simonsays1980/bayespin documentation built on Dec. 23, 2021, 2:25 a.m.