Different fees for different short holding periods

[bitcoinerS]

I was thinking some more about my above post.  I realized that, instead of pulling a table of numbers out of thin air, we can derive the curve from geometric-random-walk theoretical model of how prices change over time.  Basically you assume the current price incorporates all available information, and moves based on the cumulative effect (sum) of a bunch of small random events.  Which means the price at time t will be drawn from a Gaussian distribution with mean equal to the current price and standard deviation proportional to sqrt(t).  But this happens in log-space since investments compound, so it's actually log(price) that's Gaussian with standard deviation proportional to sqrt(t).  (This is a widely used mathematical model of prices.)

The reason we require 1x collateral is we're trying to confine the probability of a black swan [1] at the expiration time T to some bound p_swan.  I'm pretty sure if you believe the geometric-random-walk model, the log of the margin requirement should actually be proportional to sqrt(T).  Setting the 360-day requirement to be equal to the original table gives us these numbers:

Code: [Select]

duration   | min_rat
-----------|----------
14 days    | 0.31x
30 days    | 0.49x
60 days    | 0.76x
90 days    | 1.00x
120 days   | 1.23x
180 days   | 1.67x
270 days   | 2.32x
360 days   | 3.00x

[2**math.sqrt(t / 90.0)-1 for t in [14,30,60,90,120,180,270,360]]     # python code to generate above numbers

A 3.0x collateral requirement for a 1-year short implies a 90-day time horizon for 1.0x collateral.  But mathematical models are not reality, and allowing shorts with less than 1x collateral may be controversial in terms of adding black swan risk to the system.  So I suggest using this formula / table, but forcing the minimum collateral to be at least 1x.  The new table is:

Code: [Select]

duration   | min_collateral    | order_priority
-----------|-------------------|-----------------------
14 days    | 1.00x             | collateral_rat / 0.31
30 days    | 1.00x             | collateral_rat / 0.49
60 days    | 1.00x             | collateral_rat / 0.76
90 days    | 1.00x             | collateral_rat / 1.00
120 days   | 1.23x             | collateral_rat / 1.23
180 days   | 1.67x             | collateral_rat / 1.67
270 days   | 2.32x             | collateral_rat / 2.32
360 days   | 3.00x             | collateral_rat / 3.00

where of course collateral_rat is shorter_collateral / (price * quantity) and min_collateral is the minimum value of collateral_rat required for the short to execute.

[1] An adverse price movement that wipes out the collateral and puts the short "underwater," i.e. the BitUSD that must be destroyed to cover the short is more valuable than the collateral that would be unlocked.

How about targeting collateral range up from 2x?
for example:
 2x min collateral for 0+ days
 to
3x min collateral for 360 days

[theoretical]

I was thinking some more about my above post.  I realized that, instead of pulling a table of numbers out of thin air, we can derive the curve from geometric-random-walk theoretical model of how prices change over time.  Basically you assume the current price incorporates all available information, and moves based on the cumulative effect (sum) of a bunch of small random events.  Which means the price at time t will be drawn from a Gaussian distribution with mean equal to the current price and standard deviation proportional to sqrt(t).  But this happens in log-space since investments compound, so it's actually log(price) that's Gaussian with standard deviation proportional to sqrt(t).  (This is a widely used mathematical model of prices.)

The reason we require 1x collateral is we're trying to confine the probability of a black swan [1] at the expiration time T to some bound p_swan.  I'm pretty sure if you believe the geometric-random-walk model, the log of the margin requirement should actually be proportional to sqrt(T).  Setting the 360-day requirement to be equal to the original table gives us these numbers:

Code: [Select]

duration   | min_rat
-----------|----------
14 days    | 0.31x
30 days    | 0.49x
60 days    | 0.76x
90 days    | 1.00x
120 days   | 1.23x
180 days   | 1.67x
270 days   | 2.32x
360 days   | 3.00x

[2**math.sqrt(t / 90.0)-1 for t in [14,30,60,90,120,180,270,360]]     # python code to generate above numbers

A 3.0x collateral requirement for a 1-year short implies a 90-day time horizon for 1.0x collateral.  But mathematical models are not reality, and allowing shorts with less than 1x collateral may be controversial in terms of adding black swan risk to the system.  So I suggest using this formula / table, but forcing the minimum collateral to be at least 1x.  The new table is:

Code: [Select]

duration   | min_collateral    | order_priority
-----------|-------------------|-----------------------
14 days    | 1.00x             | collateral_rat / 0.31
30 days    | 1.00x             | collateral_rat / 0.49
60 days    | 1.00x             | collateral_rat / 0.76
90 days    | 1.00x             | collateral_rat / 1.00
120 days   | 1.23x             | collateral_rat / 1.23
180 days   | 1.67x             | collateral_rat / 1.67
270 days   | 2.32x             | collateral_rat / 2.32
360 days   | 3.00x             | collateral_rat / 3.00

where of course collateral_rat is shorter_collateral / (price * quantity) and min_collateral is the minimum value of collateral_rat required for the short to execute.

[1] An adverse price movement that wipes out the collateral and puts the short "underwater," i.e. the BitUSD that must be destroyed to cover the short is more valuable than the collateral that would be unlocked.

[theoretical]

I like mf-tzo's idea to allow expiry date of one year or less to be chosen by the user.

I suggest making a larger minimum requirement for collateral ratio depending on age.  E.g.:

Code: [Select]

duration   | min_rat
-----------|----------
14 days    | 1.00x
30 days    | 1.20x
60 days    | 1.30x
90 days    | 1.50x
120 days   | 1.75x
180 days   | 2.00x
270 days   | 2.50x
360 days   | 3.00x

The min_rat is the minimum collateral required to initiate the position, and also serves as a divisor for the collateral amount when ordering the short book.  E.g. the system will prefer a 14-day short at 2.01x collateral to a one-year short at 6.00x collateral (but would switch to preferring the one-year short at 6.03x collateral or better).

In other words, there are now two dimensions of user-specified parameters:  Collateral ratio and expiration date.  The system "likes" higher ratios and earlier expiration dates, but sometimes the system will have to trade off these two preferences against each other.  I.e. the system will sometimes have to choose between matching a bid against a short-term short order without very much collateral, or a long-term short order with lots of collateral.  And I'm suggesting implementing this comparison by computing offer.collateral_rat / offer.min_rat for all offers in the short wall, and putting the offer with the highest value first.

BTSX bulls can then make a choice between a long-term low-leverage bet, or short-term high-leverage bet.  The duration can be any number of blocks between 14 days and one year, the system should use some reasonable interpolation of the above table in that case (or just hard-code some polynomial function which gives approximately the desired values.)

The larger collateralization is effectively charging the long-term shorts interest in the form of forcing them to forgo the greater returns they could have gotten at greater leverage.  (The network is unable to predict future prices and determine if they would have realized greater returns or greater losses, but the short seller him/herself must believe that greater leverage would lead to greater returns [1], otherwise they wouldn't be short-selling!)

[1] Assuming that the bullish BTSX bet is a small fraction of their overall investment portfolio.

[starspirit]

keep it please as simple as possible.
No more rules please...
If possible remove some rules when we are ready to do it!

Rules like this theoretical will work but practical will not!
Traders are no bots (machines). I want to trade without needing to read a 10 page wiki (I don't want to read a single page)


NO RULES       PLEASE



PS adding more rules you will get you more liquidity from traders that choose to stay on our exchange (only our community members) BUT you will loose more liquidity from the traders they will stop trading (due your added/"advanced" rules) and because many potential future traders will never step up when they smell  complexity....

I felt the same way but I think the actual buying and selling of BitAssets is being made very simple. Very soon it will seem no different than buying and selling on bter for example. (Though I'd like to see the current yield expressed as an annual percentage and displayed clearly)

Shorting something however is different/advanced and is on an advanced screen so I don't mind if they add some rules that are clearly explained. There are no shortage of people wanting to short so I don't think it's a big problem if they add a few rules myself as long as the normal buying and selling screens are super simple.

I agree its best for trading activity to keep the buy/sell process as simple as possible, but I disagree that some shorting complexity is not an issue for them. Traders will need to use shorts if they want short or long/short (around neutral) positions.

[Empirical1.1]

keep it please as simple as possible.
No more rules please...
If possible remove some rules when we are ready to do it!

Rules like this theoretical will work but practical will not!
Traders are no bots (machines). I want to trade without needing to read a 10 page wiki (I don't want to read a single page)


NO RULES       PLEASE



PS adding more rules you will get you more liquidity from traders that choose to stay on our exchange (only our community members) BUT you will loose more liquidity from the traders they will stop trading (due your added/"advanced" rules) and because many potential future traders will never step up when they smell  complexity....

I felt the same way but I think the actual buying and selling of BitAssets is being made very simple. Very soon it will seem no different than buying and selling on bter for example. (Though I'd like to see the current yield expressed as an annual percentage and displayed clearly)

Shorting something however is different/advanced and is on an advanced screen so I don't mind if they add some rules that are clearly explained. There are no shortage of people wanting to short so I don't think it's a big problem if they add a few rules myself as long as the normal buying and selling screens are super simple.

[bytemaster]

I expected the price to rise sometime in probably the next few months, not necessarily real soon...

the goal of the proposal is to avoid such free lunch an incentive traders to make more short/medium term predictions.

Almost feels like there should be a "grandfathering" clause or something then, I would've never opened the shorts if I would have known that. I've already lost some BTSX getting margin called in the BitCNY market (due to the low liquidity, not the actual price on the exchange) and I wouldn't have risked more of my balance if I didn't think I could hold it until BTSX really took off.

 

I agree.  It doesn't seem fair to apply this to existing shorts.

I agree with grandfathering if we adopt such a policy.

[liondani]

keep it please as simple as possible.
No more rules please...
If possible remove some rules when we are ready to do it!

Rules like this theoretical will work but practical will not!
Traders are no bots (machines). I want to trade without needing to read a 10 page wiki (I don't want to read a single page)


NO RULES       PLEASE



PS adding more rules you will get you more liquidity from traders that choose to stay on our exchange (only our community members) BUT you will loose more liquidity from the traders they will stop trading (due your added/"advanced" rules) and because many potential future traders will never step up when they smell  complexity....

[mf-tzo]

the reason I am still on the fence though is not because I don't believe that BTSX will be worth billions but rather because I don't know how to compete with trading bots, the GUI is not stable, I don't know yet when a margin call happens so I don't want to risk too much of my BTSX, I don't know why when I place a short order at the bid price is not executed, I don't know a lot of things yet which make me feel uncomfortable...

Imposing fees and expiry restrictions at this time I think will create further confusions.. Let's trade everyone as we can with the already complicated issues and decisions that each one of us face before complicate things even further...

Other than that, it absolutely make sense a short order to have an expiry. But maybe this expiry should not be imposed by the system but rather from the user. I think it would make sense to treat that like forward contracts are treated.

eg. You open a short position and depending on the chosen expiry you have to pay some additional fees according to the forward curve on this expiry date. But how can you create a forward curve with bitassets and charge fair fees in such an iliquid market? That's why for now I wouldn't further complicate things...

[santaclause102]

please do not impose fees and short expiration days to the shorts...I am having difficulty in deciding to short or not already. If fees are imposed then liquidity may be more reduced.

An expiration day of 6m - 1y maybe should be ok though and would make sense

It is a good thing (for the overall system) that you are on the fence. Liquidity should be about equal on both sides.

[mf-tzo]

please do not impose fees and short expiration days to the shorts...I am having difficulty in deciding to short or not already. If fees are imposed then liquidity may be more reduced.

An expiration day of 6m - 1y maybe should be ok though and would make sense

[santaclause102]

I expected the price to rise sometime in probably the next few months, not necessarily real soon...

the goal of the proposal is to avoid such free lunch an incentive traders to make more short/medium term predictions.

what is the "free lunch" here? long term risk for long term potential gain as well as potential loss.
an efficient market needs a cross-section of traders, from short to medium term, trend vs value etc. homogeneity means no need for the market, no trading.

It is easy to predict that btsx will go up in price at some point in the future. This contributes to the shorts overhang. The proposal would also incentivize traders to trade more medium term which will help track the price more accurately.

[pendragon3]

I expected the price to rise sometime in probably the next few months, not necessarily real soon...

the goal of the proposal is to avoid such free lunch an incentive traders to make more short/medium term predictions.

Almost feels like there should be a "grandfathering" clause or something then, I would've never opened the shorts if I would have known that. I've already lost some BTSX getting margin called in the BitCNY market (due to the low liquidity, not the actual price on the exchange) and I wouldn't have risked more of my balance if I didn't think I could hold it until BTSX really took off.

 

I agree.  It doesn't seem fair to apply this to existing shorts.

Agree. Existing short positions should be grandfathered in. Otherwise, it wouldn't be fair for those who took a risk to invest in their long-term views under the publicized rules that were in force at the time. What happened to the dictum that "everyone should get what they ask for" and not be forced to do something against their will?

More generally, forcing shorts to cover periodically doesn't seem like a good idea. It's a heavy-handed measure that is unlikely to increase liquidity much if at all. The reason is that it disincentivizes would-be short sellers too much. A short-seller who is forced to cover won't be able to easily re-short, as they have to compete with numerous others on collateral amount to carry out another short sale. The uncertainty of whether rolling into another short is feasible is disruptive to medium-term or long-term trading strategies of the bulls. It's an attempt to create greater balance between bulls and bears, but at a high cost and in an inefficient way.

Why not have shorts pay a modest amount of interest over time? (The rate should be modest and capped, perhaps tied to inflation). This way, shorts can choose how long to hold their positions. This would be like what happens in traditional markets with short selling of equity.

[GaltReport]

I expected the price to rise sometime in probably the next few months, not necessarily real soon...

the goal of the proposal is to avoid such free lunch an incentive traders to make more short/medium term predictions.

Almost feels like there should be a "grandfathering" clause or something then, I would've never opened the shorts if I would have known that. I've already lost some BTSX getting margin called in the BitCNY market (due to the low liquidity, not the actual price on the exchange) and I wouldn't have risked more of my balance if I didn't think I could hold it until BTSX really took off.

 

I agree.  It doesn't seem fair to apply this to existing shorts.

[Empirical1.1]

I think I'm in favour of long term shorts, people who want to hold longer than a month competing via fees/interest rate they're willing to pay after a certain max collateral is reached.

Something like the bidding system here maybe, 

https://bitsharestalk.org/index.php?topic=9364.0

The interest rate acts as the price. The trades are conducted using the "price" and time as the two major factors.
Price at transaction: The offers provided by the borrowing party with a higher price take precedence over those with a lower price.
Time: When the two parties have the same offered price, the bids put in first take precedence over those offered later. The transactions will follow chronological order going from those submitted first to last. 

This lets people be long term shorts without punishing market makers and raises funds to add to BitYield.

(There should be lots of churn because people will offer high interest rates and max collateral to get the short and then just cover within a month to avoid the interest rate and if they don't BitYield wins and more BitAsset creation is incentivised.)

[Markus]

The risk with punishing shorts is that you break the peg the other way. If there were a lack of shorts what is going to keep the BitAsset prices from becoming overvalued?

Introducing a fee shorts have to pay for long-term holding is equivalent to the interest discussion (the thing) we had half a year ago.