Sample Size

From Open Risk Manual
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Definition

Sample Size is an important consideration in the development of risk models based on historical data. It concerns the total number of distinct observations that are available for estimating a model, with the dimensions such as the length of temporal observation window or the number of entities governing the sample size

Minimal Sample Size

Depending on the nature of the model (implicit or explicit distribution assumptions, number of parameters etc.) the required sample size will vary significantly. Minimal Sample Size is an estimated minimum number that would ensure the estimation is statistically sound