Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - bitcrab

Pages: 1 ... 68 69 70 71 72 73 74 [75] 76 77 78 79 80 81 82 ... 129
1111
actually BSIP42 redefined "feed price" as something like "guide price",  not the exact market price.

there's someways to get smartcoin premium, for bitCNY, you can

1. get the bitCNY deposit/withdraw fee from magicwallet and then calculate out the bitCNY premium.

or

2. from exchange like AEX or ZB get BTS price in CNC or QC, it can be regarded as BTS price in fiat as CNC and QC are IOU of CNY.  with the BTS price in bitCNY, you can calculate out the premium, 

1112
Who is the judge of what the correct fiat price is?
How did you calculate the 9% from the 2%? Could you elaborate that please?

logic:

1. 2% premium means bitCNY is already in serious shortage, need strong negative feedback.
2. when feed price < DEX price*MSSR, it is always safe as the margin called orders can always be eaten.

MSSR = 1.1, to be careful, we select 9%, not 10% as the metric.

1113
SPRING management team is considering to vote witnesses to push the implementation of BSIP42 if the voting WP is accepted at 7th Sep.

we'd like to adopt one metric to judge whether one witness work well on implementing BSIP42: if bitCNY has a more than 2% premium than fiat, then the fed price should be at least 9% higher than BTS price in DEX. currently only bitCNY feed price is considered.

if SPRING holders have any thoughts on this plan, please comment here.

1114
调喂价和调MCR没有本质区别,只是后者看起来更直观而已。
喂价与内盘价之比高过MSSR,会明显增加黑天鹅的风险,这一点毋庸赘述。

1115
基于abit的这个算法可以实现快速纠偏
Quote
喂价 = BTS市场价 * (bitcny/cny价)^1.8

这种幂函数算法的缺点很明显,就是溢价不是太高时改进效果不明显。

比如当溢价=2%时,喂价=BTS市场价*1.036

而其实溢价2%已经是很严重的通缩了,应该采取足够大的反馈力度了,这时完全可以喂价=BTS市场价*1.08,保证爆仓单可以被吃的安全边际条件下最大化喂价。

不是到很严重的下跌市才需要大力度的反馈,当前也需要!


1116
负反馈未必能完全解决溢价问题,但可以为解决问题创造条件。

把喂价提升到Pdex*MSSR的范围内是很安全的,之外会有一些风险,那就至少先在很安全的范围内充分发挥负反馈的力量吧。

1117
负反馈的力度还都太低了。
溢价大于2%的时候,喂价好歹比内盘价高个8%以上吧。

1118
建议以后可以增加“喂价质量评估”栏目,方便投票者评估见证人喂价质量。尤其是实施负反馈的质量 :)

1119
gdex-witness新的喂价算法:

Pdex:内盘bitCNY价格
premium:溢价百分比

if 0<premium<2%:
feed price = Pdex*(1+4.8*premium)

if 2%<premium<4.8%:
feed price = Pdex*(1+9.6%)

if 4.8%<premium:
feed price = Pdex*(1+2*premium)

思路:当溢价已经比较严重(>2%)时,直接将喂价设为最高安全价(爆仓时爆仓单依然可以被吃掉的喂价),溢价低于2%时,线性下降保证溢价为0时喂价下降到内盘价,溢价更严重时(>4.8%),按“溢价多少补多少”的原则继续线性增加补偿,不设上限。

预计今天上线,继续观察效果。
 

1120

我现在喂价用的公式是

喂价 = BTS市场价 * (bitcny/cny价)^1.8

市场价用的是全场交易量加权平均,zb、coinegg因为刷量明显所以降权了。

修正系数用的是1.8次方,而不仅仅是加上溢价。这个值可以调,之前试过1.5感觉偏小、用2感觉偏大,继续观察看看。

没有设上限,也就是说bitcny溢价越大,喂价调整比例越大。

不错,有创意。

1121
世界不是你想象的这个样子的。
说我贪婪/激进,我没意见。
但abit贪婪吗?大山贪婪吗?他们在这次熊市中也被爆得很惨。
你说市场跌太多就应该增加抵押率,问题在于,熊市中谁有能力随时拿出足够多的资金来增加抵押率?
按你的规则,结果不是比特股获得强劲动力,而是没人愿意抵押了。

1122
Stakeholder Proposals / Re: [Poll] BSIP42: adjust price feed dynamically
« on: August 29, 2018, 09:00:53 am »
https://github.com/bitshares/bsips/pull/106

add discussion and voting relevant content.

Did you see my post above? I did the exact same thing. Shall we merge the PRs?

oh, sorry I haven't checked your PR details when I write my PR, now seems my PR can be ignored...

1123
So the buyback of SPRING will be at the same price in BTS terms, but you are expecting BTS to rise compared to Yuan (and overall)? At least if the fund is successful.

Yes, we hope BTS ecosystem can grow, and also the BTS market cap.

1124
Stakeholder Proposals / Re: [Poll] BSIP42: adjust price feed dynamically
« on: August 29, 2018, 07:21:27 am »
https://github.com/bitshares/bsips/pull/106

add discussion and voting relevant content.

1125
如果有效果就持续下去吧,暂停的代价太大。。。

Pages: 1 ... 68 69 70 71 72 73 74 [75] 76 77 78 79 80 81 82 ... 129