Detection of Stock Price Manipulation Using Kernel Based Principal Component Analysis and Multivariate Density Estimation

Stock price manipulation uses illegitimate means to artificially influence market prices of several stocks. It causes massive losses and undermines investors’ confidence and the integrity of the stock market. Several existing research works focused on detecting a specific manipulation scheme using supervised learning but lacks the adaptive capability to capture different manipulative strategies. This begets the assumption of model parameter values specific to the underlying manipulation scheme. In addition, supervised learning requires the use of labelled data which is difficult to acquire due to confidentiality and the proprietary nature of trading data. The proposed research establishes a detection model based on unsupervised learning using Kernel Principal Component Analysis (KPCA) and applied increased variance of selected latent features in higher dimensions. A proposed Multidimensional Kernel Density Estimation (MKDE) clustering is then applied upon the selected components to identify abnormal patterns of manipulation in data. This research has an advantage over the existing methods in overcoming the ambiguity of assuming values of several parameters, reducing the high dimensions obtained from conventional KPCA and thereby reducing computational complexity. The robustness of the detection model has also been evaluated when two or more manipulative activities occur within a short duration of each other and by varying the window length of the dataset fed to the model. Validation on multiple datasets and a comprehensive assessment of the model performance has been conducted without providing any prior information about the location of the manipulation. The results show a significant performance enhancement in terms of the F-measure values and a significant reduction in false alarm rate (FAR) has been achieved.

[1]  Tom C. W. Lin.  The New Market Manipulation , 2017 .

[2]  Franklin Allen,et al.  Stock Price Manipulation, Market Microstructure and Asymmetric Information , 1991 .

[3]  P. Bonissone,et al.  Fuzzy ROC curves for unsupervised nonparametric ensemble techniques , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[4]  Ammar Belatreche,et al.  A Dendritic Cell Immune System Inspired Approach for Stock Market Manipulation Detection , 2019, 2019 IEEE Congress on Evolutionary Computation (CEC).

[5]  P. J. Green,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[6]  Li Liao,et al.  Phylogenetic tree information aids supervised learning for predicting protein-protein interaction based on distance matrices , 2007, BMC Bioinformatics.

[7]  Wiro J. Niessen,et al.  Quantitative Comparison of Spot Detection Methods in Fluorescence Microscopy , 2010, IEEE Transactions on Medical Imaging.

[8]  Dell Zhang.  Detecting Network Anomalies With Kernel Principal Component Analysis , 2006 .

[9]  Akira Maeda,et al.  Unsupervised Outlier Detection in Time Series Data , 2006, 22nd International Conference on Data Engineering Workshops (ICDEW'06).

[10]  Eun Jung Leea,et al.  Microstructure-based manipulation: Strategic behavior and performance of spoofing traders , 2013 .

[11]  Abhijit Guha Roy,et al.  DASA: Domain adaptation in stacked autoencoders using systematic dropout , 2015, 2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR).

[12]  Wei Xu,et al.  Enhancing intraday stock price manipulation detection by leveraging recurrent neural networks with ensemble learning , 2019, Neurocomputing.

[13]  Craig Pirrong.  The economics of commodity market manipulation: A survey , 2017 .

[14]  Ammar Belatreche,et al.  Adaptive Hidden Markov Model With Anomaly States for Price Manipulation Detection , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[15]  S. Rhee,et al.  Trade-based manipulation: Beyond the prosecuted cases , 2017 .

[16]  Bart De Moor,et al.  New bandwidth selection criterion for Kernel PCA: Approach to dimensionality reduction and classification problems , 2014, BMC Bioinformatics.

[17]  Ruihong Huang,et al.  The Market Impact of a Limit Order , 2011 .

[18]  L. C. Matioli,et al.  A new algorithm for clustering based on kernel density estimation , 2018 .

[19]  M. C. Jones,et al.  A Comparison of Higher-Order Bias Kernel Density Estimators , 1997 .

[20]  Emmanuel Haven,et al.  De-noising option prices with the wavelet method , 2012, Eur. J. Oper. Res..

[21]  Franklin Allen,et al.  Stock-Price Manipulation , 1992 .

[22]  Domenico Giannone,et al.  Economic Predictions with Big Data: The Illusion of Sparsity , 2017, Econometrica.

[23]  Philip S. Yu,et al.  Coupled Behavior Analysis with Applications , 2012, IEEE Transactions on Knowledge and Data Engineering.

[24]  Yuxin Ding,et al.  Host-based intrusion detection using dynamic and static behavioral models , 2003, Pattern Recognit..

[25]  Osmar R. Zaïane,et al.  Detecting stock market manipulation using supervised learning algorithms , 2014, 2014 International Conference on Data Science and Advanced Analytics (DSAA).

[26]  Bertrand B. Maillet,et al.  Outliers Detection, Correction of Financial Time-Series Anomalies and Distributional Timing for Robust Efficient Higher-Order Moment Asset Allocations , 2009 .

[27]  Farshad Moradi,et al.  Pain detection from facial images using unsupervised feature learning approach , 2016, 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[28]  Heiko Hoffmann,et al.  Kernel PCA for novelty detection , 2007, Pattern Recognit..

[29]  VARUN CHANDOLA,et al.  Anomaly detection: A survey , 2009, CSUR.

[30]  I. Johnstone,et al.  Ideal spatial adaptation by wavelet shrinkage , 1994 .

[31]  M. Hazelton.  Variable kernel density estimation , 2003 .

[32]  Weihua Li,et al.  Recursive PCA for adaptive process monitoring , 1999 .

[33]  Babis Theodoulidis,et al.  Financial Markets Monitoring and Surveillance: A Quote Stuffing Case Study , 2012 .

[34]  F. Alt,et al.  Choosing principal components for multivariate statistical process control , 1996 .

[35]  Ammar Belatreche,et al.  An experimental evaluation of novelty detection methods , 2014, Neurocomputing.

[36]  Bernhard Schölkopf,et al.  Sparse Kernel Feature Analysis , 2002 .

[37]  S. X. Yang,et al.  An Adaptive Approach Based on KPCA and SVM for Real-Time Fault Diagnosis of HVCBs , 2011, IEEE Transactions on Power Delivery.

[38]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[39]  So Young Sohn,et al.  Stock fraud detection using peer group analysis , 2012, Expert Syst. Appl..

[40]  Ji Wu,et al.  Maximum Margin Clustering Based Statistical VAD With Multiple Observation Compound Feature , 2011, IEEE Signal Processing Letters.

[41]  Richard Payne,et al.  Computer-Based Trading and Market Abuse , 2012 .

[42]  Heng Tao Shen,et al.  Principal Component Analysis , 2009, Encyclopedia of Biometrics.

[43]  Meng-Han Yang,et al.  Discrimination of China's stock price manipulation based on primary component analysis , 2014, 2014 International Conference on Behavioral, Economic, and Socio-Cultural Computing (BESC2014).

[44]  Theodora Kourti,et al.  Process analysis, monitoring and diagnosis, using multivariate projection methods , 1995 .

[45]  Tom Fawcett,et al.  An introduction to ROC analysis , 2006, Pattern Recognit. Lett..

[46]  Ronald M. Summers,et al.  Optimizing area under the ROC curve using semi-supervised learning , 2015, Pattern Recognit..

[47]  Babis Theodoulidis,et al.  Analysis of stock market manipulations using knowledge discovery techniques applied to intraday trade prices , 2011, Expert Syst. Appl..

[48]  Amir F. Atiya,et al.  Introduction to the special issue on neural networks in financial engineering , 2001, IEEE Trans. Neural Networks.

[49]  M. C. Jones,et al.  Comparison of Smoothing Parameterizations in Bivariate Kernel Density Estimation , 1993 .

[50]  Quan Wang,et al.  Kernel Principal Component Analysis and its Applications in Face Recognition and Active Shape Models , 2012, ArXiv.

[51]  Y. Kwok.  Mathematical models of financial derivatives , 2008 .

[52]  Dirk P. Kroese,et al.  Kernel density estimation via diffusion , 2010, 1011.2602.

[53]  Aristides Gionis,et al.  k-means-: A Unified Approach to Clustering and Outlier Detection , 2013, SDM.

[54]  Suttipong Thajchayapong,et al.  Stock price manipulation detection using a computational neural network model , 2016, 2016 Eighth International Conference on Advanced Computational Intelligence (ICACI).

[55]  Ramazan Aktas,et al.  Detecting stock-price manipulation in an emerging market: The case of Turkey , 2009, Expert Syst. Appl..