Theory-Driven model goodness of fit test on Cryptocurrency Price Data

Journal of Advanced Technology Research, Vol. 3, No. 1, pp. 7-14, Jun. 2018
10.11111/JATR.2018.3.1.007, Full Text:
Keywords: Cryptocurrency, Price Prediction, Regression test, Correlation Analysis
Abstract

The stock price prediction has been tried many times, but stock price is a time series data, so it is difficult to predict its price. There are many methodology related to stock price prediction but predicting exact price of stock is impossible. Starting with Bitcoin, the market for cryptocurrency is getting more active now, and efforts to predict cryptocurrency prices are being made. Cryptocurrency has characteristics that are similar to stock price data, but it is more volatile and random. We collect a 12 cryptocurrency data through the Bithumb API, and pre-process it. Finally, we analyze the correlation between cryptocurrency pairs to check to apply the regression model, which is a conventional theory- driven stock price prediction model. From this analysis, we argue that the theory-driven model is not proper to predict the cryptocurrency prices.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
D. Kwon, J. Heo, J. Kim and Y. Han, "Theory-Driven model goodness of fit test on Cryptocurrency Price Data," Journal of Advanced Technology Research, vol. 3, no. 1, pp. 7-14, 2018. DOI: 10.11111/JATR.2018.3.1.007.

[ACM Style]
Do-Hyung Kwon, Ju-Seong Heo, Ju-Bong Kim, and Youn-Hee Han. 2018. Theory-Driven model goodness of fit test on Cryptocurrency Price Data. Journal of Advanced Technology Research, 3, 1, (2018), 7-14. DOI: 10.11111/JATR.2018.3.1.007.