New report available, updated results

A new report is available. Subscribe to the newsletter to receive the full report. People donating to this project will receive the full report, along with the upcoming ICO report.

Attached I made a short analysis of the returns that the top 10 coins made in the reports that were sent out last week and two weeks ago. The results are promising. However please note that these results do not in any way guarantee results in the future. Moreover, markets have been good and there is not enough data to test for significance. Use this report as a signal and always do your own research before investing. No rights can be derived of this and I am not responsible for errors in the calculation.

20171129 Valuationreport_nc

New report available

A new report is available and has been send to subscribers. For those interested, the following updates are in the works:

  • ICO sentiment analysis
  • Adding of StockTwits data
  • Overall crypto market sentiment indicators
  • Finetuning of the “drop alert” indicator (indicator of decreasing sentiment of coins)
  • Quantitative regression methods in which Crypto’s are most sensitive of Social Media sentiment

20171125 Valuationreport

An example of Twitter frontrunning price

Growing the dataset gives opportunity to do some more detailed analysis of looking in to sentiment metrics vs prices. In the picture I took a look at LSK for which I took a position a while ago based on rising number of recipients. The chart does give some indication of a leading relationship. Although the dataset has to grow in order to make statistically significant statement, this does looks promising in my opinion.

New valuation scorecard available! Further improvements underway

I’ve created a new valuation scorecard which will be added to the regular report for  the coming days. See below for a detailed explanation.

20171114 Valuationreport

I am a believer of fundamental (not technical) analysis in order to value coins (as a supplement to plain research and common sense). Coming from a traditional investment background I coded a scorecard to value cryptocurrency based on Twitter, Google, Reddit and Github analysis . The code score the metrics below with a Z-score (each value is score relative to a normal distribution with mean = 0 and standard deviation = 1. Then, given pre determined weights, each coin is scored on how it compares to the other top 100 coins. In case someone has feedback or questions let me know

  • Twitter metrics
    • Mentions – Number of Tweets on a coin relative to its market cap
    • Recipients – Number of people that receive the Tweet relative to its market cap
    • Sentiment – Average tone of Tweets (positive or negative)
  • Reddit Metrics
    • Subscr – Numbers of subscribers to a subreddit relative to market cap
    • U onl – Number of average online users on a Subreddit relative to market cap
    • Sentiment – Average tone of posts of top 25 (positive or negative)
  • Google Trends
    • Number of searches on coin relative to its market cap
  • Github
    • Avg commits over last month
    • Number of additions over last month