Quantcast
Channel: Active questions tagged r - Stack Overflow
Viewing all articles
Browse latest Browse all 201919

Factor Analysis Between Cryptocurrencies and Fiat Currencies

$
0
0

I am currently working on a project that requires me to determine whether or not cryptocurrencies may be used as a global currency. I am working with a dataset which includes variables such as bitcoin, ethereum, litecoin, Chinese Yuan (fiat), Euro (fiat), Pound (fiat), gold prices, crude oil prices, and several indices such as Dow Jones, S&P, NYSE ext. I have filtered my data so that it looks something like this:

Bitcoin Ethereum Litecoin    Gold   Oil Chinese Yuan Europe  Pound      SP Dow Jones EPU Index
  576.31    12.37     3.86 1305.06 43.17       6.6807 0.8988 0.7618 2173.56  18430.05     38.52
  575.64    12.41     3.93 1308.71 44.39       6.6832 0.8971 0.7548 2184.87  18544.76     40.85
  610.83    11.86     3.97 1322.60 44.85       6.6903 0.8979 0.7521 2186.57  18551.54     30.28
  614.54    11.72     4.03 1346.43 45.47       6.6751 0.8907 0.7509 2187.87  18536.89     22.17
  628.77    11.76     3.99 1342.44 47.63       6.6691 0.8902 0.7530 2184.94  18506.24     27.04
  626.83    12.06     3.99 1333.20 45.88       6.6842 0.8932 0.7554 2169.08  18404.17     35.65
     NASDAQ New York Stock Exchange Volatil Index
 5229.94                10780.71           156
 5263.39                10867.85            48
 5275.91                10891.22            15
 5287.61                10898.76            36
 5271.01                10903.86            79
 5225.93                10799.13           468

I have explored the correlation between these variables and have seen that the cryptocurrencies have strong positive correlation among themselves, but seem to have a negative correlation with fiat currencies. This leads me to suspect that cryptocurrencies can't be considered a global currency, but I would like to use some sort of statistical technique to back up/disprove my claim. I have conducted a factor analysis on subset of this dataset which only includes the fiat and the cryputocurrencies. The results are as follows:

Uniquenesses:
     Bitcoin     Ethereum     Litecoin Chinese Yuan       Europe        Pound 
       0.100        0.104        0.100        0.190        0.109        0.100 

Loadings:
             Factor1 Factor2
Bitcoin      -0.429   0.853 
Ethereum     -0.566   0.759 
Litecoin     -0.457   0.845 
Chinese Yuan  0.802  -0.408 
Europe        0.823  -0.462 
Pound         0.846  -0.437 

               Factor1 Factor2
SS loadings      2.750   2.590
Proportion Var   0.458   0.432
Cumulative Var   0.458   0.890

Test of the hypothesis that 2 factors are sufficient.
The chi square statistic is 194.34 on 4 degrees of freedom.
The p-value is 6.19e-41 

As you can see here, there is a clear distinction between fiat currencies and cryptocurrencies in the loadings. Is this sufficient evidence to conclude that cryptocurrencies cannot be considered a global currencies? If not, then could you please point me in the right direction? Are there any other techniques I should consider?


Viewing all articles
Browse latest Browse all 201919

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>