simulate_ekop_mis: Simulates mis-specified trades data

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

Description

Simulates mis-specified trades data from the model by Easley, Kiefer, O'Hara and Paperman (1996). The simulation function is implemented in C++ and allows therefore usage in specific performance-sensitive scenarios, like simulating huge amounts of trade data or calling this function repeatedly in large simulations.

Usage

simulate_ekop_mis(
  size = 1000,
  alpha = 0.2,
  epsilon = 0.2,
  delta = 0.5,
  mu = 0.02,
  T = 60 * 6.5,
  mis = 0.15
)

Arguments

size	An integer specifying the number of trading days that should be simulated.
alpha	A double specifying the probability of an information event.
epsilon	A double specifying the arrival rate of uninformed traders.
delta	A double specifying the probability of information being good or bad.
mu	A double specifying the arrival rate of informed traders.
T	A double defining the length of a trading day in minutes.
mis	A double specifying the mis-sepcification rate for trades.

Details

The functions returns a data.frame with named columns MisBuy, MisSell, Buy, Sell, and Trades. Buy and Sell are the number of buy and sell trades per trading day and Trades defines the sum of these amounts. MisBuy and MisSell define the number of mis-specified buy and sell trades per trading day respectively.

Value

A data.frame containing the trades per day.

References

• Easley, D., Kiefer, N., O’Hara, M., Paperman, J., 1996. Liquidity, information, and infrequently traded stocks. Journal of Finance 51, 1405–1436.

• Grammig, J., Theissen, E., Zehnder, L.S., 2015. Bayesian Estimation of the Probability of Informed Trading. Conference on Financial Econometrics & Empirical Asset Pricing 2016, Lancaster University

See Also

• simulate_ekop() for simulating trades data without mis-specification

• correct_trades() to reconstruct original trades data from mis-specified ones

Examples

# Simulate mis-specifed trades data for 1000 trading dates with a trading 
# day lasting 6.5 hours.
trades_data <- simulate_ekop_mis(size = 1000, alpha = .3, epsilon = .4,
                                 delta = .5, mu = .04, T = 60*6.5, mis = .1)
head(trades_data)

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