Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ
ÇѱÛÁ¦¸ñ(Korean Title) |
¾ÏÈ£ÈÆó Á¾°¡ ¿¹Ãø ¼º´É°ú ÀÔ·Â º¯¼ö °£ÀÇ ¿¬°ü¼º ºÐ¼® |
¿µ¹®Á¦¸ñ(English Title) |
Understanding the Association Between Cryptocurrency Price Predictive Performance and Input Features |
ÀúÀÚ(Author) |
¹ÚÀçÇö
¼¿µ¼®
Jaehyun Park
Yeong-Seok Seo
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 11 NO. 01 PP. 0019 ~ 0028 (2022. 01) |
Çѱ۳»¿ë (Korean Abstract) |
ÃÖ±Ù ¾ÏÈ£ÈÆó°¡ ¸¹Àº ÁÖ¸ñÀ» ¹ÞÀ½¿¡ µû¶ó ¾ÏÈ£ÈÆóÀÇ Á¾°¡ ¿¹Ãø ¿¬±¸µéÀÌ È°¹ßÈ÷ ÁøÇàµÇ°í ÀÖ´Ù. ƯÈ÷ µö ·¯´× ¸ðµ¨À» Àû¿ë½ÃÄÑ ¿¹Ãø ¼º´ÉÀ» ³ôÀÌ·Á´Â ¿¬±¸µéÀÌ Áö¼ÓµÇ°í ÀÖ´Ù. µö ·¯´× ¸ðµ¨ Áß ½Ã°è¿ µ¥ÀÌÅÍ¿¡¼ ³ôÀº ¿¹Ãø ¼º´ÉÀ» º¸ÀÌ´Â LSTM (Long Short-Term Memory) ¸ðµ¨ÀÌ ´Ù°¢µµ·Î ÀÀ¿ëµÇ°í ÀÖÀ¸³ª º¯µ¿¼ºÀÌ Å« ¾ÏÈ£ÈÆó Á¾°¡ µ¥ÀÌÅÍ¿¡¼´Â ³·Àº ¿¹Ãø ¼º´ÉÀ» º¸ÀδÙ. À̸¦ ÇØ°áÇϱâ À§ÇØ »õ·Î¿î ÀÔ·Â º¯¼ö¸¦ ã¾Æ³»°í, À̸¦ »ç¿ëÇÏ´Â Á¾°¡ ¿¹Ãø ¿¬±¸°¡ ¼öÇàµÇ°í ÀÖ´Ù. ±×·¯³ª µö ·¯´× ±â¹ÝÀÇ ¾ÏÈ£ÈÆó Á¾°¡ ¿¹Ãø¿¡ »ç¿ëµÇ´Â µ¥ÀÌÅ͵éÀÇ °¢ ÀÔ·Â º¯¼öµéÀÌ ¿¹Ãø ¼º´É¿¡ ¹ÌÄ¡´Â ¿µÇâ·ÂÀ̳ª ÇнÀ¿¡ È¿À²ÀûÀÎ ÀÔ·Â º¯¼öµéÀÇ Á¶ÇÕ¿¡ °üÇÑ ¿¬±¸ »ç·Ê°¡ ºÎÁ·ÇÑ ½ÇÁ¤ÀÌ´Ù. µû¶ó¼ º» ³í¹®¿¡¼´Â Bitcoin°ú EthereumÀ» Æ÷ÇÔÇÑ 6°¡Áö ¾ÏÈ£ÈÆóÀÇ ÃÖ±Ù µ¿Çâ ÀڷḦ ¼öÁýÇÏ¿´°í, Åë°è¿Í µö ·¯´×À» ÅëÇØ ÀÔ·Â º¯¼öµéÀÌ ¾ÏÈ£ÈÆó Á¾°¡ ¿¹Ãø¿¡ ¹ÌÄ¡´Â ¿µÇâ·ÂÀ» ºÐ¼®ÇÑ´Ù. ½ÇÇè °á°ú ¸ðµç ¾ÏÈ£ÈÆóÀÇ Á¾°¡ ¿¹Ãø ¼º´É Æò°¡¿¡¼ Á¾°¡ º¯µ¿·üÀ» Á¦¿ÜÇÑ °³Àå°¡, °í°¡, Àú°¡, °Å·¡·®, Á¾°¡¸¦ Á¶ÇÕÇßÀ» ¶§ °¡Àå ¿ì¼öÇÑ ¼º´ÉÀ» º¸¿´´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Recently, cryptocurrency has attracted much attention, and price prediction studies of cryptocurrency have been actively conducted. Especially, efforts to improve the prediction performance by applying the deep learning model are continuing. LSTM (Long Short-Term Memory) model, which shows high performance in time series data among deep learning models, is applied in various views. However, it shows low performance in cryptocurrency price data with high volatility. Although, to solve this problem, new input features were found and study was conducted using them, there is a lack of study on input features that drop predictive performance. Thus, in this paper, we collect the recent trends of six cryptocurrencies including Bitcoin and Ethereum and analyze effects of input features on the cryptocurrency price predictive performance through statistics and deep learning. The results of the experiment showed that cryptocurrency price predictive performance the best when open price, high price, low price, volume and price were combined except for rate of closing price fluctuation. |
Å°¿öµå(Keyword) |
LSTM
Deep Learning
ÀÔ·Â º¯¼ö
¾ÏÈ£ÈÆó
°¡°Ý ¿¹Ãø
µ¥ÀÌÅÍ ºÐ¼®
LSTM
Deep Learning
Input Feature
Cryptocurrency
Price Prediction
Data Analysis
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|