Skip to main content

Table 5 Relative comparison with similar studies

From: Multivariate cryptocurrency prediction: comparative analysis of three recurrent neural networks approaches

Authors (year)

Data source

Cryptocurrencies

Methods

Avg. MAPE (best)

Radityo et al. [16]

cryptocompare.com

Bitcoin (BTC)

BPNN

GANN

GABPNN

NEAT

1.998

4.461

1.883

2.175

Patel et al. [2]

investing.com

Litecoin (LTC)

Monero (XMR)

LSTM (LTC)

LSTM (XMR)

GRU-LSTM (LTC)a

GRU-LSTM (XMR)a

6.3109

5.4405

2.0581

6.2754

Jay et al. [24]

coinmarketcap.com

Bitcoin (BTC)

Ethereum (ETH)

Litecoin (LTC)

Stoc. MLP (BTC)

Stoc. MLP (ETH)

Stoc. MLP (LTC)

Stoc. LSTM (BTC)

Stoc. LSTM (ETH)

Stoc. LSTM (LTC)

2.96007

2.55175

2.47681

3.19191

3.46424

2.69210

Karova et al. [38]

CoinAPI

Bitcoin (BTC)

LSTM

3.353

Buyrukoglu [39]

coinmarketcap.com

Audius (AUDI)

Bitcoin Gold (BTG)

OKB

Pax Dollar (USDP)

Telcoin (TEL)

LSTM (AUDI)

LSTM (BTG)

LSTM (OKB)

LSTM (USDP)

LSTM (TEL)

Ens. LSTM (AUDI)a

Ens. LSTM (BTG)a

Ens. LSTM (OKB)a

Ens. LSTM (USDP)a

Ens. LSTM (TEL)a

0.43

0.78

0.44

0.09

0.83

0.27

0.75

0.69

0.12

0.52

Our approach

finance.yahoo.com

Bitcoin (BTC)

Ethereum (ETH)

Cardano (ADA)

Tether (USDT)

Binance Coin (BNB)

LSTM

Bi-LSTM

GRU

0.052991

0.0465712

0.0446512

  1. aThe proposed method in those studies