Market Predictability - page 13

[John Seekers]

Geometric phases describe how in a continuous-time dynamical system the displacement of a variable (called phase variable) can be related to other variables (shape variables) undergoing a cyclic motion, according to an area rule. The aim of this paper is to show that geometric phases can exist also for discrete-time systems, and even when the cycles in shape space have zero area. A context in which this principle can be applied is stock trading. A zero-area cycle in shape space represents the type of trading operations normally carried out by high-frequency traders (entering and exiting a position on a fast time-scale), while the phase variable represents the cash balance of a trader. Under the assumption that trading impacts stock prices, even zero-area cyclic trading operations can induce geometric phases, i.e., profits or losses, without affecting the stock quote.

[John Seekers]

We discuss Bayesian forecasting of increasingly high-dimensional time series, a key area of application of stochastic dynamic models in the financial industry and allied areas of business. Novel state-space models characterizing sparse patterns of dependence among multiple time series extend existing multivariate volatility models to enable scaling to higher numbers of individual time series. The theory of these "dynamic dependence network" models shows how the individual series can be "decoupled" for sequential analysis, and then "recoupled" for applied forecasting and decision analysis. Decoupling allows fast, efficient analysis of each of the series in individual univariate models that are linked-- for later recoupling-- through a theoretical multivariate volatility structure defined by a sparse underlying graphical model. Computational advances are especially significant in connection with model uncertainty about the sparsity patterns among series that define this graphical model; Bayesian model averaging using discounting of historical information builds substantially on this computational advance. An extensive, detailed case study showcases the use of these models, and the improvements in forecasting and financial portfolio investment decisions that are achievable. Using a long series of daily international currency, stock indices and commodity prices, the case study includes evaluations of multi-day forecasts and Bayesian portfolio analysis with a variety of practical utility functions, as well as comparisons against commodity trading advisor benchmarks.

[John Seekers]

Shang and Hyndman (2016) proposed grouped functional time series forecasting approach as a combination of individual forecasts using generalized least squares regression. We modify their methodology using generalized exponential smoothing technique for the most disaggregated series in order to obtain more robust predictor. We show some properties of our proposals using simulations and real data related to electricity demand prediction.

[John Seekers]

We propose a new method (implemented in an R-program) to simulate long-range daily stock-price data. The program reproduces various stylized facts much better than various parametric models from the extended GARCH-family. In particular, the empirically observed changes in unconditional variance are truthfully mirrored in the simulated data.

[John Seekers]

In this article, we examine the recent shift in financial markets toward high-frequency trading (HFT). This turn is being legitimized with reference to how algorithms are allegedly more rational and efficient than human traders, and less prone to emotionally motivated decisions. We argue that although HFT does not render humans irrelevant, it is leading to a reconfiguration of both the ideal trading subject and the human–machine relations. Drawing on interviews with and ethnographic observations of high-frequency traders, as well as HFT ‘how to’ books, we analyze the subjectivity and self-techniques of the ideal high-frequency trader. We demonstrate that these traders face the challenge of avoiding emotional interference in their algorithms and that they deploy a set of disciplinary self-techniques to curb the importance of emotional attachment.

[John Seekers]

There exists vast research articles which predict the stock market as well pricing of stock index financial instruments but most of the proposed models focus on the accurate forecasting of the levels (i.e. value) of the underlying stock index. There is a lack of studies examining the predictability of the direction/sign of stock index movement. Given the notion that a prediction with little forecast error does not necessarily translate into capital gain, this study is an attempt to predict the direction of S&P CNX NIFTY Market Index of the National Stock Exchange, one of the fastest growing financial exchanges in developing Asian countries. Random forest and Support Vector Machines (SVM) are very specific type of machine learning method, and are promising tools for the prediction of financial time series. The tested classification models, which predict direction, include linear discriminant analysis, logit, artificial neural network, random forest and SVM. Empirical experimentation suggests that the SVM outperforms the other classification methods in terms of predicting the direction of the stock market movement and random forest method outperforms neural network, discriminant analysis and logit model used in this study.