- See Section 2.379.
- Response parameter is pseudo-acceleration for 5% damping.
- Only use data within passband of high-pass filtering fhp ≤ 1∕(1.25T) hence number of records per T
- Based on distribution of fhp with Mw, conclude that some noise may still be present in records from small
events. Also note large event term for 2004 Baladeh (Iran, Mw6.2) that relate to poor filtering.
- Coefficient b3 is not significantly different than zero for 0.15–1.5s and c3 is only significantly different than
zero for 0.04–0.4s.
- Sharp drop in σ for T > 3s, which believe is not reliable. Hence do not report coefficients for T > 3s.