Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market
The purpose of this research is to add to the literature on the topic of market efficiency for the cryptocurrency market, adding to the previous findings through testing for weak-form efficiency of the returns between bullish and bearish samples. In addition to this, the study attempts to accurately...
| Main Author: | |
|---|---|
| Format: | Dissertation (University of Nottingham only) |
| Language: | English |
| Published: |
2021
|
| Online Access: | https://eprints.nottingham.ac.uk/66443/ |
| _version_ | 1848800329526673408 |
|---|---|
| author | Morris, Jack |
| author_facet | Morris, Jack |
| author_sort | Morris, Jack |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | The purpose of this research is to add to the literature on the topic of market efficiency for the cryptocurrency market, adding to the previous findings through testing for weak-form efficiency of the returns between bullish and bearish samples. In addition to this, the study attempts to accurately forecast the volatility of the returns, which is already well regarded as being greater than other traditional financial markets. This was achieved through the use of a Generalised Autoregressive Conditional Heteroskedasticity (1,1) model. Five cryptocurrencies were analysed based on their market capitalisation, ignoring stable and meme coins. These coins were Bitcoin, Ethereum, Ripple, Cardano and Stellar. With three samples for each taken, a rolling-window which is regarded as the entire population data, then a collective bull cycle sample and a collective bear cycle sample, derived from the rolling window population data. From the GARCH (1,1) model, I was able to distinguish volatility clusters being common in the market as well as volatility being persistent. In addition to this, from the Portmanteau test for white noise, I was able to add to the previous conflicting literature by concluding that the market, from the population data, is inefficient. This is beneficial to investors in the market due to their technical analysis being of use to gain excessive returns on the cryptocurrencies that are presently undervalued because of a time lag in the returns. My investigation also added to the efficiency topic of the cryptocurrency market by being able to state that bearish periods are more efficient than bullish periods. Therefore, the cycle the market is within does impact players in the markets possibility at seeking excess returns due to the predictability power varying between cycles. |
| first_indexed | 2025-11-14T20:49:50Z |
| format | Dissertation (University of Nottingham only) |
| id | nottingham-66443 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:49:50Z |
| publishDate | 2021 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-664432023-04-25T09:48:22Z https://eprints.nottingham.ac.uk/66443/ Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market Morris, Jack The purpose of this research is to add to the literature on the topic of market efficiency for the cryptocurrency market, adding to the previous findings through testing for weak-form efficiency of the returns between bullish and bearish samples. In addition to this, the study attempts to accurately forecast the volatility of the returns, which is already well regarded as being greater than other traditional financial markets. This was achieved through the use of a Generalised Autoregressive Conditional Heteroskedasticity (1,1) model. Five cryptocurrencies were analysed based on their market capitalisation, ignoring stable and meme coins. These coins were Bitcoin, Ethereum, Ripple, Cardano and Stellar. With three samples for each taken, a rolling-window which is regarded as the entire population data, then a collective bull cycle sample and a collective bear cycle sample, derived from the rolling window population data. From the GARCH (1,1) model, I was able to distinguish volatility clusters being common in the market as well as volatility being persistent. In addition to this, from the Portmanteau test for white noise, I was able to add to the previous conflicting literature by concluding that the market, from the population data, is inefficient. This is beneficial to investors in the market due to their technical analysis being of use to gain excessive returns on the cryptocurrencies that are presently undervalued because of a time lag in the returns. My investigation also added to the efficiency topic of the cryptocurrency market by being able to state that bearish periods are more efficient than bullish periods. Therefore, the cycle the market is within does impact players in the markets possibility at seeking excess returns due to the predictability power varying between cycles. 2021-09-09 Dissertation (University of Nottingham only) NonPeerReviewed application/pdf en https://eprints.nottingham.ac.uk/66443/1/Master%20Dissertation%20pdf.pdf Morris, Jack (2021) Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market. [Dissertation (University of Nottingham only)] |
| spellingShingle | Morris, Jack Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market |
| title | Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market |
| title_full | Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market |
| title_fullStr | Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market |
| title_full_unstemmed | Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market |
| title_short | Volatility forecasting in Bull and Bear Cycles: Evidence from the Cryptocurrency Market |
| title_sort | volatility forecasting in bull and bear cycles: evidence from the cryptocurrency market |
| url | https://eprints.nottingham.ac.uk/66443/ |