Forecasting Power of Sentiment on Stock Return
(2015) BUSP70 20151Department of Business Administration
- Abstract
- We study the predictability of sentiment on stock return of the Hong Kong market. We measure sentiment by forming composite sentiment indices from several proxies using principal component analysis and the partial least square method. We find that sentiment measures using either method have similar in-sample forecasting power but the one formed by using partial least square method has better out-of-sample predictability on stock return. We do not find any sentiment predictability on the cross-section of stock return.
- Popular Abstract
- We study the predictability of sentiment on stock return of the Hong Kong market. We measure sentiment by forming composite sentiment indices from several proxies using principal component analysis and the partial least square method. We find that sentiment measures using either method have similar in-sample forecasting power but the one formed by using partial least square method has better out-of-sample predictability on stock return. We do not find any sentiment predictability on the cross-section of stock return.
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/5468904
- author
- Pong, King Yin LU and Cao, Yu LU
- supervisor
- organization
- course
- BUSP70 20151
- year
- 2015
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- principal component analysis, return, sentiment, partial least square
- language
- English
- id
- 5468904
- date added to LUP
- 2015-06-08 16:00:25
- date last changed
- 2015-06-08 16:00:25
@misc{5468904, abstract = {{We study the predictability of sentiment on stock return of the Hong Kong market. We measure sentiment by forming composite sentiment indices from several proxies using principal component analysis and the partial least square method. We find that sentiment measures using either method have similar in-sample forecasting power but the one formed by using partial least square method has better out-of-sample predictability on stock return. We do not find any sentiment predictability on the cross-section of stock return.}}, author = {{Pong, King Yin and Cao, Yu}}, language = {{eng}}, note = {{Student Paper}}, title = {{Forecasting Power of Sentiment on Stock Return}}, year = {{2015}}, }