InsidersTradesSigma
Performance & Risk Metrics
A stochastic computational method that estimates Value at Risk by generating thousands of random market scenarios based on estimated price distributions and correlations, used in quant scoring to assess tail risk exposure across insider-trading-flagged portfolios.
Monte Carlo VaR simulates thousands of possible future price paths under assumed volatilities and correlations, then reads the loss quantile at the chosen confidence level (typically 95% or 99%). Unlike parametric VaR, it makes no normality assumption, so it can handle non-linear payoffs such as options, fat tails, and changing correlations. The trade-off is computational cost and sensitivity to the assumptions you feed in.
The quality of the output depends entirely on the inputs: the volatility and correlation estimates, and the assumed distribution of shocks. Garbage in, garbage out applies fully here. Because it samples randomly, the estimate also carries simulation error that shrinks only as you run more paths, so report enough iterations for the figure to stabilise.
Formula