Abstract Randomness is a key assumption in various important financial and economic theories. Therefore, the assessment of randomness in data holds pivotal significance in the effective application of these theories. However, very few tests are available to check for randomness across diverse scenarios. This paper introduces a new approach using the concept of random interval graphs (RIG) to create a non-parametric test for evaluating randomness in data. Our approach is based on the idea that the properties of a RIG are independent of the choice of distribution of observations and, therefore work for a wide range of application setups. We also validate the effectiveness of the test through various simulation experiments and real-world data. |
Pstujeme web | visit: Skluzavky