Cryptocurrency Survival: An Analysis

of the Price and Future of Bitcoin

Avinash Ramesh

Overview

• Price History• Volatility Analysis• Price Drivers• Short-term Predictive Techniques• Projections for 2015• Long-term Survival and Projections

Price History – General Overview

Price History 2014 – A Year in Review

• Price went from $763.97/BTC to $313.92/BTC• Ross Ulbricht on trial for running Silk Road marketplace• January 9th – Overstock.com announced it would accept

bitcoin• February 18th – First bitcoin ATM opened in the US• February 25th – Mt. Gox shutdown• February – Slovenian Exchange BitStamp experiences DDOS

attack• December 19th – Charlie Shrem sentenced to two years jail

for helping Silk Road customers illegally launder money through BitInstant

Price Drivers• Transaction Volume

• The amount of bitcoin moved around the system

• Supply• Lower than 13 million due to lost bitcoin

• Expectations of Future Value• Uncertainty of Future Value

• Leads to volatility• High uncertainty/volatility bad because

bitcoin should be a store of value

“While the eMail pointing to a $50,000 bitcoin is imprecise to say the least, It’s a starting point. A more rigorous mathematical model integrating demand projections with probabilities of certain make-or-break contingencies isn’t just possible – it already exists, in various versions, across various investors’ hard drives.”

Volatility Analysis

Volatility Analysis (cont.)

8/16

/201

0

9/27

/201

0

11/8

/201

0

12/2

0/20

10

1/31

/201

1

3/14

/201

1

4/25

/201

1

6/6/

2011

7/18

/201

1

8/29

/201

1

10/1

0/20

11

11/2

1/20

11

1/2/

2012

2/13

/201

2

3/26

/201

2

5/7/

2012

6/18

/201

2

7/30

/201

2

9/10

/201

2

10/2

2/20

12

12/3

/201

2

1/14

/201

3

2/25

/201

3

4/8/

2013

5/20

/201

3

7/1/

2013

8/12

/201

3

9/23

/201

3

11/4

/201

3

12/1

6/20

13

1/27

/201

4

3/10

/201

4

4/21

/201

4

6/2/

2014

7/14

/201

4

8/25

/201

4

10/6

/201

4

11/1

7/20

14

12/2

9/20

14

2/9/

2015

3/23

/201

5

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

f(x) = − 1.06597932790227 ln(x) + 11.3840809186819R² = 0.126920684689609

Volatility

Volatility Analysis (Cont.)

8/16

/201

0

9/20

/201

0

10/2

5/20

10

11/2

9/20

10

1/3/

2011

2/7/

2011

3/14

/201

1

4/18

/201

1

5/23

/201

1

6/27

/201

1

8/1/

2011

9/5/

2011

10/1

0/20

11

11/1

4/20

11

12/1

9/20

11

1/23

/201

2

2/27

/201

2

4/2/

2012

5/7/

2012

6/11

/201

2

7/16

/201

2

8/20

/201

2

9/24

/201

2

10/2

9/20

12

12/3

/201

2

1/7/

2013

2/11

/201

3

3/18

/201

3

4/22

/201

3

5/27

/201

3

7/1/

2013

8/5/

2013

9/9/

2013

10/1

4/20

13

11/1

8/20

13

12/2

3/20

13

1/27

/201

4

3/3/

2014

4/7/

2014

5/12

/201

4

6/16

/201

4

7/21

/201

4

8/25

/201

4

9/29

/201

4

11/3

/201

4

12/8

/201

4

1/12

/201

5

2/16

/201

5

3/23

/201

5

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

Volatility of Volatility

Price Drivers - News

Price Drivers (cont.)• Velocity• Number of times a unit of currency is used in a fixed period

of time• Low velocity -> higher price

• Liquidity• Hoarders’ coins are illiquid

• Human Emotion• Can’t Model

• Lower bound – mining reward must be greater than mining costs

Short-term Predictive Techniques - Overview

• Spot Price

• Bayesian Regression• Paper by Shah and Zhang• Assumes technical methods can be used to predict future price• Uses nonparametric regression• Trial period in paper returned 89% under model

• Artificial Neural Networks• Commonly used for pattern recognition and nonlinear curve

fitting

Short-Term Predictive Techniques - Problems

• Bayesian Regression• Trial period was only 60 days• Did not take into account transaction fees• Gain was over a timespan in which bitcoin price rose

dramatically• Appears inconclusive

• Artificial Neural Networks• Too many hidden-layers leads to overfitting, too few leads to

underfitting

Projections for 2015 – The Good

• Boston Retail Partners believe 8% of retailers will accept Bitcoin• This year to see integration of BitLicense, which sets

rules and regulations for BTC merchants• CoinBase announced the release of an insured

exchange in February• Major companies such as Microsoft, Overstock.com,

and Dell now take Bitcoin

Projections for 2015 – The Bad

• According to “The Guardian”, 65% of general public still doesn’t understand Bitcoin• Of those with awareness of Bitcoin, 85% have never

used it• Juniper Research predicts transactions for

cryptocurrencies to fall as much as 50%• Transactions from 2014 artificially inflated due to advent of

altcoins like Dogecoin

Long-Term Survival Critics and Responses

• Gold cannot be compared to Bitcoin due to intrinsic value of gold• Response: gold’s intrinsic value is quite low, both Bitcoin and gold

are general stores of value

• Adoption of near-perfect substitute AltCoins creates a potentially unlimited amount of BitCoin• Response: Network Effects

• Lack of government control• Governments can impose taxes in fiat currencies which prevents

full adoption of Bitcoin• Response A: Government control will play a larger factor in

Bitcoin’s future• Response B: Bitcoin is meant to be decentralized

Long-Term Survival Critiques and Responses• Most people holding it are speculating• True, but move from 95%/5% to 80%/20% speculation to

payment ratio.

• Several illegal transactions• According to Fred Ehrsam, cofounder of CoinBase, SilkRoad

transactions account for <1% of total

Long-Term Survival Projection: S-Curve

• Current Problem of Chicken or Egg

• Waiting for critical mass of merchants and consumers

Long-Term Survival Projections: Network Effects

• Bitcoin will reach a critical mass before altcoins, after which its network effect will keep it as the dominant cryptocurrency

• Comparison of Bitcoin to Internet

Bitcoin Price Simulator• Price = (TransactionVol / Velocity) * (1 / Supply)• http://

worldbitcoinnetwork.com/BitcoinPriceModel-Alpha.html

Conclusion• 2014 was a rough year for Bitcoin in terms of price,

but also showed long-term potential• Analysis of Bitcoin’s volatility shows that it has been

steadily decreasing• Bitcoin is still in the early adoption phase of the S

curve, and has yet to reach a critical mass of people• Bitcoin’s price is complex and affected by several

factors; however, there exist models to try to account for these factors

Sources• http://arxiv.org/pdf/1410.1231v1.pdf• https://willyreport.wordpress.com/2014/05/25/the-willy-report-proof-of-massive-fraudulent-tr

ading-activity-at-mt-gox-and-how-it-has-affected-the-price-of-bitcoin/• http://www.paymentlawadvisor.com/files/2014/01/GoldmanSachs-Bit-Coin.pdf• http://www.newsbtc.com/2015/02/13/juniper-predicts-future-bitcoin-will-two-years/• http://www.btcpredictions.com/index.html• http://www.theguardian.com/technology/2015/feb/03/bitcoin-2015-make-or-break-year• http://www.newsbtc.com/2015/02/05/report-predicts-50-drop-cryptocurrency-transactions-ye

ar/• http://www.coindesk.com/juniper-research-bitcoin-transactions-double-2017/• http://www.newsbtc.com/2014/12/31/microsoft-experimenting-bitcoin/• https://www.cryptocoinsnews.com/predicting-bitcoin-value-50000/• https://www.youtube.com/watch?v=g2nXgK34HIM&feature=share• http://insidebitcoins.com/news/bitcoin-2014-growing-adoption-high-profile-arrests-and-a-bew

ildering-price/28060• http://www.cnbc.com/id/102301308• https://www.cryptocoinsnews.com/affects-bitcoin-price/• https://jmarcedwards.files.wordpress.com/2011/01/networkeffect1.jpg