α = 2.00  1 Month (30d)  6 Months (180d)  1 Year (365d) 

Performance  {{ (t.r.month >=0 ? '+' :'') }}{{ t.r.month  number: 2 }} %  {{ (t.r['6month'] >=0 ? '+' :'') }}{{ t.r['6month']  number: 2 }} %  {{ (t.r['year'] >=0 ? '+' :'') }}{{ t.r['year']  number: 2}} % 
High  {{ t.h.month * 1000  number:0 }}  {{ t.h['6month'] * 1000  number:0 }}  {{ t.h['year'] * 1000  number:0 }} 
Low  {{ t.l.month * 1000  number:0 }}  {{ t.l['6month'] * 1000  number:0 }}  {{ t.l['year'] * 1000  number:0 }} 
Volatility  {{ t.v.month  number: 2 }} %  {{ t.v['6month']  number: 2 }} %  {{ t.v['year']  number: 2}} % 
TokenTicker

Market Cap (USD M) at {{ scopeMapped.chartData.biccxHistoricValue[scopeMapped.chartData.biccxHistoricValue.length 1].seriesDate  date:'dd.MM.yyyy' }}  Price at {{ scopeMapped.chartData.biccxHistoricValue[scopeMapped.chartData.biccxHistoricValue.length 1].seriesDate  date:'dd.MM.yyyy' }}  Daily Price Change 
TypeFocus


{{ val.fullname }}{{ key }}

{{ val.lastMarketCap / 1000000  number: 2 }}  ${{ val.lastPrice  number: 2 }}  {{ (val.lastPriceChange >=0 ? '+' :'') }}{{ val.lastPriceChange * 100  number: 2 }}% 
{{ val.tokentype }}{{ val.focus }}

We define \(\operatorname{M}({c,t})\) as the market capitalization of crypto asset \(c\) at time \(t\).
The classic weighting by market capitalization is given by
\[w_{c,t} = \frac{\operatorname{M}({c,t})}{\sum\limits_{k\in C}\operatorname{M}({k,t})}\text{.}\]
The problem with this approach becomes visible by looking at the current market capitalizations of the crypto asset market.
Bitcoin dominates the general crypto asset market and also our selection of assets.
To fight the dominance of Bitcoin we would like to reduce the proportion of large crypto assets and shift over some weight to smaller ones.
Therefore we define a function that has a decreasing gradient.
This function will relatively increase small values and decrease big values.
In our case we choose
\[f(x)=x^{\frac{1}{\alpha}}\text{.}\]
We call \(\alpha\) the redistribution factor.
By applying \(f\) to \(w\) we shift the weights.
\[
w_\text{new}=\frac{f(w)}{\sum_{i} f(w_i)}
\]
By choosing an \(\alpha\) you can refine the degree of redistribution.
An \(\alpha\) of 1 implies \(f(x)=x\), meaning the transformation won't have any effect.
This represents the base case where we choose the weights by their proportional market capitalization.
An \(\alpha>1\) will overweigh the assets with relatively small market caps.
When \(\alpha=\infty\) this portfolio is equal to the \(1/N\)Portfolio.
An \(0<\alpha<1\) will overweigh assets that already have a huger market cap.
An negative \(\alpha\) will flip the roles and the smallest assets become the biggest ones.
The weighting function is then given by
\[
w_{c,t} = \frac{f(\operatorname{M}({c,t}))}{\sum\limits_{k\in C}f(\operatorname{M}({k,t}))} = \frac{\operatorname{M}({c,t})^{\frac{1}{\alpha}}}{\sum\limits_{k\in C}\operatorname{M}({k,t})^{\frac{1}{\alpha}}}\text{.}
\]
Below we attached an animation of how the weights behave for different redistribution values of \(\alpha\).
Every investor may choose his own \(\alpha\). BICCX uses an default \(\alpha\) value of 2.
You can also smooth the market cap using a rolling window mean
\[
\hat{\operatorname{M}}({c,t}) = \frac{1}{T} \sum_{i=0}^{T1}\operatorname{M}({c, ti})
\]
This will reduce variance in weight changes. We provide an option to smooth the market cap by using an 7 day mean \((T=7)\).
The value of the BICCX is calculated as
\[
I(t)=\sum_{c\in C} w_{c,t} \frac{P({c,t})}{P({c,T_{c,0}})}
\]
where \(T_{c,0}\) corresponds to the 01.07.2018 and \(P({c,t})\) is the price of coin \(c\) at time \(t\).
The index is calculated once a day with data recorded at 00:00 UTC
CoinMetrics is a leading crypto financial data provider for institutions. The company emerged as the trusted brand for building institutionalgrade crypto asset infrastructure. Besides, it provides valuation frameworks for investors and empowers them to better understand, value, use, and ultimately steward public crypto networks.
Didier Goepfert  Project Lead
Didier is specialized in investments and new product/market launch in the financial services industry. He has shared the journey with several successful and innovative startups such as Raisin (online savings marketplace), Machinio (search engine) or Streamr (blockchainpowered data marketplace) navigating in various industries and countries. Didier graduated from ESMT Berlin (MBA) and ClermontFerrand University (MSc in Financial Markets).
Paulo Morales Castillo  Data Lead
At Kapilendo, Paulo oversees the company’s data strategy, including the implementation of database solutions, prediction algorithms, and reporting systems in the organization. Before joining Kapilendo, he was a senior data analyst at the creditcomparison site Smava, where he worked for 4 years. Paulo earned bachelor’s degrees in mathematics and economics from the California State University at Sacramento, and an M.S. in economics from the University of Bonn.