The Hz ABA oscillates 14% which produces predictable peak/trough phases - the effect of these peak/troughs on the market activity/behaviour could be observed for the relevant price feed premium effect you're talking about.
Initially the amplitude was set to 50%, reduced to 33% then finally set to 14% due to concerns of undesirable volatility/velocity, however there's been low market activity thus far - 14% may not be enough. There's been times where I've shorted the peak of Hz, only for BTS to change 30% the next day wiping out my upcoming benefits of shorting Hz (28% debt reduction), perhaps instead it'd be more attractive to account for the 30d BTS:USD (or BTS:CNY in your case) price volatility (via an ES query) to create a varying amplitude which may be more attractive (however I don't think you'd want a -+50% oscillation, right?)..
Would you consider applying a small oscillation to the price feed of the MPA in question? You could then create a second identical MPA with a 180' phase offset to mirror the oscillation of the primary MPA (or even 3 with 120' phase offset from one another).
If you're talking about MSSR & MCR values set in price feeds, then I've always just copied what the committee have configured & I'm unsure of the benefits/drawbacks of changing these values - it probably should be configured by the asset owner instead of the price feed publishers though, no? Plus these fields need better explanations om the docs like scenarios where it'd be configured high or low & effects that'd have on the MPA's operation.