Virtual Library of Simulation Experiments:

Screening Test Problems

---

Linkletter et al. (2006) No Signal Function

Description:

Dimensions: 10

This function is used for input variable screening, to see how well a method can identify a lack of signal. As mentioned in Linkletter et al. (2006), Welch (1992) pointed out that it is difficult to distinguish between a model with no active factors and one with all active factors. A small error term ε ~ N(0, 0.05²) is added to the response.

Input Domain:

This function is evaluated on the scaled input domain x_i ∈ [0, 1], for all i = 1, …, 10.

Code:

References:

Linkletter, C., Bingham, D., Hengartner, N., Higdon, D., & Kenny, Q. Y. (2006). Variable selection for Gaussian process models in computer experiments. Technometrics, 48(4).

Welch, W. J., Buck, R. J., Sacks, J., Wynn, H. P., Mitchell, T. J., & Morris, M. D. (1992). Screening, predicting, and computer experiments. Technometrics, 34(1), 15-25.

---

For questions or comments, please email Derek Bingham at: dbingham@stat.sfu.ca.