Forecasting multinomial stock returns using machine learning methods
In this paper, the daily returns of the S&P 500 stock market index are predicted using a variety of different machine learning methods. We propose a new multinomial classification approach to forecasting stock returns. The multinomial approach can isolate the noisy fluctuation around zero return...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2020-11-01
|
| Series: | Journal of Finance and Data Science |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405918820300143 |
| id |
doaj-7eaa5a58acde4e068a909f1c9ebbecfb |
|---|---|
| record_format |
Article |
| spelling |
doaj-7eaa5a58acde4e068a909f1c9ebbecfb2021-04-02T19:14:23ZengKeAi Communications Co., Ltd.Journal of Finance and Data Science2405-91882020-11-01686106Forecasting multinomial stock returns using machine learning methodsLauri Nevasalmi0Department of Economics, Turku School of Economics, University of Turku, FI-20014, Finland; Department of Mathematics and Statistics, University of Turku, FI-20014, FinlandIn this paper, the daily returns of the S&P 500 stock market index are predicted using a variety of different machine learning methods. We propose a new multinomial classification approach to forecasting stock returns. The multinomial approach can isolate the noisy fluctuation around zero return and allows us to focus on predicting the more informative large absolute returns. Our in-sample and out-of-sample forecasting results indicate significant return predictability from a statistical point of view. Moreover, all the machine learning methods considered outperform the benchmark buy-and-hold strategy in a real-life trading simulation. The gradient boosting machine is the top-performer in terms of both the statistical and economic evaluation criteria.http://www.sciencedirect.com/science/article/pii/S2405918820300143S&P 500Market timingMachine learningGradient boosting |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Lauri Nevasalmi |
| spellingShingle |
Lauri Nevasalmi Forecasting multinomial stock returns using machine learning methods Journal of Finance and Data Science S&P 500 Market timing Machine learning Gradient boosting |
| author_facet |
Lauri Nevasalmi |
| author_sort |
Lauri Nevasalmi |
| title |
Forecasting multinomial stock returns using machine learning methods |
| title_short |
Forecasting multinomial stock returns using machine learning methods |
| title_full |
Forecasting multinomial stock returns using machine learning methods |
| title_fullStr |
Forecasting multinomial stock returns using machine learning methods |
| title_full_unstemmed |
Forecasting multinomial stock returns using machine learning methods |
| title_sort |
forecasting multinomial stock returns using machine learning methods |
| publisher |
KeAi Communications Co., Ltd. |
| series |
Journal of Finance and Data Science |
| issn |
2405-9188 |
| publishDate |
2020-11-01 |
| description |
In this paper, the daily returns of the S&P 500 stock market index are predicted using a variety of different machine learning methods. We propose a new multinomial classification approach to forecasting stock returns. The multinomial approach can isolate the noisy fluctuation around zero return and allows us to focus on predicting the more informative large absolute returns. Our in-sample and out-of-sample forecasting results indicate significant return predictability from a statistical point of view. Moreover, all the machine learning methods considered outperform the benchmark buy-and-hold strategy in a real-life trading simulation. The gradient boosting machine is the top-performer in terms of both the statistical and economic evaluation criteria. |
| topic |
S&P 500 Market timing Machine learning Gradient boosting |
| url |
http://www.sciencedirect.com/science/article/pii/S2405918820300143 |
| work_keys_str_mv |
AT laurinevasalmi forecastingmultinomialstockreturnsusingmachinelearningmethods |
| _version_ |
1721549284690624512 |