Learning Connections in Financial Time Series

Gartheeban Ganeshapillai, John Guttag, Andrew Lo
Proceedings of the 30th International Conference on Machine Learning, PMLR 28(2):109-117, 2013.

Abstract

To reduce risk, investors seek assets that have high expected return and are unlikely to move in tandem. Correlation measures are generally used to quantify the connections between equities. The 2008 financial crisis, and its aftermath, demonstrated the need for a better way to quantify these connections. We present a machine learning-based method to build a connectedness matrix to address the shortcomings of correlation in capturing events such as large losses. Our method uses an unconstrained optimization to learn this matrix, while ensuring that the resulting matrix is positive semi-definite. We show that this matrix can be used to build portfolios that not only “beat the market,” but also outperform optimal (i.e., minimum variance) portfolios.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-ganeshapillai13, title = {Learning Connections in Financial Time Series}, author = {Ganeshapillai, Gartheeban and Guttag, John and Lo, Andrew}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {109--117}, year = {2013}, editor = {Dasgupta, Sanjoy and McAllester, David}, volume = {28}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/ganeshapillai13.pdf}, url = {https://proceedings.mlr.press/v28/ganeshapillai13.html}, abstract = {To reduce risk, investors seek assets that have high expected return and are unlikely to move in tandem. Correlation measures are generally used to quantify the connections between equities. The 2008 financial crisis, and its aftermath, demonstrated the need for a better way to quantify these connections. We present a machine learning-based method to build a connectedness matrix to address the shortcomings of correlation in capturing events such as large losses. Our method uses an unconstrained optimization to learn this matrix, while ensuring that the resulting matrix is positive semi-definite. We show that this matrix can be used to build portfolios that not only “beat the market,” but also outperform optimal (i.e., minimum variance) portfolios.} }
Endnote
%0 Conference Paper %T Learning Connections in Financial Time Series %A Gartheeban Ganeshapillai %A John Guttag %A Andrew Lo %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-ganeshapillai13 %I PMLR %P 109--117 %U https://proceedings.mlr.press/v28/ganeshapillai13.html %V 28 %N 2 %X To reduce risk, investors seek assets that have high expected return and are unlikely to move in tandem. Correlation measures are generally used to quantify the connections between equities. The 2008 financial crisis, and its aftermath, demonstrated the need for a better way to quantify these connections. We present a machine learning-based method to build a connectedness matrix to address the shortcomings of correlation in capturing events such as large losses. Our method uses an unconstrained optimization to learn this matrix, while ensuring that the resulting matrix is positive semi-definite. We show that this matrix can be used to build portfolios that not only “beat the market,” but also outperform optimal (i.e., minimum variance) portfolios.
RIS
TY - CPAPER TI - Learning Connections in Financial Time Series AU - Gartheeban Ganeshapillai AU - John Guttag AU - Andrew Lo BT - Proceedings of the 30th International Conference on Machine Learning DA - 2013/05/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-ganeshapillai13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 28 IS - 2 SP - 109 EP - 117 L1 - http://proceedings.mlr.press/v28/ganeshapillai13.pdf UR - https://proceedings.mlr.press/v28/ganeshapillai13.html AB - To reduce risk, investors seek assets that have high expected return and are unlikely to move in tandem. Correlation measures are generally used to quantify the connections between equities. The 2008 financial crisis, and its aftermath, demonstrated the need for a better way to quantify these connections. We present a machine learning-based method to build a connectedness matrix to address the shortcomings of correlation in capturing events such as large losses. Our method uses an unconstrained optimization to learn this matrix, while ensuring that the resulting matrix is positive semi-definite. We show that this matrix can be used to build portfolios that not only “beat the market,” but also outperform optimal (i.e., minimum variance) portfolios. ER -
APA
Ganeshapillai, G., Guttag, J. & Lo, A.. (2013). Learning Connections in Financial Time Series. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research 28(2):109-117 Available from https://proceedings.mlr.press/v28/ganeshapillai13.html.

Related Material