Pricing of GHS and GHS Futures on CEX.IO

[arivar]

Hi all,

I've been working in a pricing model for the GHS and Futures of GHS traded in CEX.IO. Here I am going to explain the steps I followed to develop my model:

1) Estimate the future btc flow for 1 GH/s.
       For this step 3 things are needed: 1) Estimate the difficulty change over the time; 2) Estimate the mining margin; and 3) Estimate the transaction fees in the future.
2) Create a discount curve to calculate the present value of the future btc flow.


The Difficulty estimation is the most important and also the most subjective step. I believe we should be very conservative here and consider that it is going to grow at very high rates in the next two years and after this we should consider Moore's Law.
The mining margin nowadays in CEX.IO is about 86.3%, due many reasons I decided to keep it constant for the future in my model(I can explain the reasons in another post later).
To estimate the transaction fees in the future I couldn't find a better guess than using the number we have today(about 0.1btc p/ block) and increase this number proportional to the existing number of BTC in each date.

Finally, to find the fair price of the GHS we just need the discount curve. I believe a fair discount curve would be the one that keeps my share of the total existing BTC constant over time. Another thing we can do is to find the discount curve using the current market price of the GHS(goal seek), this is going to be needed for pricing future contracts.

One interest thing I decided to do about the discount curve is to use blocks instead of dates(contrary to financial market standards). Although, in the model I consider that one block takes always 10 minutes to be found, which will be almost true on average.

The model above is useful to decide if it is worth or not to buy GHS on CEX.IO.

Now, to decide if it is worth to buy the future instead of the spot GHS we should use the same model to calculate how many GHS would you have on the delivery date if you just buy GHS now, let them mine for you and buy more GHS with the revenues. By knowing this number, the decision to buy the future or the spot is quite easy.

Here are my results during the block 291358:

Current GHS Price: 0.01302
GHS Fair Price using the model: 0.01308
Current Future April Price(Block 296735): 0.00802
Future April Fair Price: 0.00991
Current Future May Price(Block 301054): 0.00410
Future May Fair Price: 0.00831

These results mean that the best decision now is to buy the future for May, because if you start mining now with 1 GHS you will have 1.57 GHS on the delivery date, so you can spend up to 0.00831 per GHS future now. With the model I also found a return of 105.27% in BTC if you buy the May future now at 0.0041.


The reason I opened this topic is to start a discussion about pricing of all kind of derivatives related to crypto-currencies, since everything is extremely new we can help each other a lot by sharing knowledge.
I made an excel spreadsheet with the model I described above (https://dl.dropboxusercontent.com/u/69035552/BTC_Pricing.xlsm), before using it remember to set your excel for manual calculation(if you let automatic calculation you excel will probably crashes). I would recomend to never hit F9, instead you should always use the button "Refresh" on tab "Pricing". Remember as well to manually update the blue cells(C3, D16 and E16) on tab Pricing, that should be enough. Fell free to use the spreadsheet and please let me know if you find any mistakes.

I will keep updating this post with updates in this model and also models for any other derivatives that interests me. If you like the work I would really appreciate donations of any size

BTC: 18TL3yWPuFkvhP4zbvQ7tjABpXDDU1Pzrf
LTC: LP25S1YLmArmpHj8bEria4wTyLKL7e6rM1

Best Regards

[1714976482]

1714976482

[1714976482]

1714976482

[greenlion]

The ability of Cex themselves to arbitrarily issue new GH/s onto the market dwarfs the predictive value of any analysis whatsoever. These are not futures or derivatives, it's a rigged monopsony market.

[arivar]

I dont see any problem with them issuing new GH/s, it should affect the Difficulty which is included in the model.

[greenlion]

It directly moves the price itself in a way that no model can predict, it's not through a proxy like difficulty.

[arivar]

Actually, it seems to be good for the holders of GHS when they increase their mining power, because maybe they can dilute the fixed costs and our mining margin could improve.

What the model predicts is the price you should pay to have an specific return on your BTCs as long as you hold the GHS. The model considers the GHS as a bond where you have a future cash flow in BTC.

[greenlion]

Actually, it seems to be good for the holders of GHS when they increase their mining power, because maybe they can dilute the fixed costs and our mining margin could improve.

What the model predicts is the price you should pay to have an specific return on your BTCs as long as you hold the GHS. The model considers the GHS as a bond where you have a future cash flow in BTC.

The fact that Cex is a monopsony GH/s supplier means that the bond has perpetually negative interest rate unless you're operating under fairytale difficulty increase projections.

When Cex dumps GH/s on the market, they are not just dilluting fixed costs, they're also dilluting GH/s price vis-a-vis simple supply-and-demand.

People don't actually make any money on Cex treating it like your model does, the future discounted revenue aspect of Cex cannot ever be profitable anytime soon, the only way to make a return is day-trading GH/s purely speculatively, which is a losing proposition because again the monopsony nature of the market means that exactly 1 participant has 100% perfect analytics and can rent-seek against the order book.

[arivar]

What do you call a "fairytale difficulty increase projections" ? CEX is increasing the supply of GH/s at a rate that is less than 10% a month(check how much they are selling), my difficult projection is more than the double of this rate.

[arivar]

I decided to update the difficulty forecast to a much more conservative projection. Now I am considering about 18% of increase in difficulty every 2016 blocks until the end of 2015, then I set a lower pace. With this very conservative scenario, the only derivative that is worth to buy is the future for may FHM.
The model tells me that this future is being sold 20% cheaper than the fair price.

[arivar]

I've changed the model a bit to price the futures and I am also considering a much more aggressive difficulty increase over the time(about 20% every 2016 blocks). It still shows a very good ROI on the future for may. If anyone wants the updated model here it is https://dl.dropboxusercontent.com/u/69035552/BTC_Pricing.xlsm.