scholarly journals An Empirical Study on Importance of Modeling Parameters and Trading Volume-Based Features in Daily Stock Trading Using Neural Networks

Informatics ◽  
2018 ◽  
Vol 5 (3) ◽  
pp. 36 ◽  
Author(s):  
Thuy-An Dinh ◽  
Yung-Keun Kwon
Keyword(s):  
Stock Price ◽  
Trading Volume ◽  
Support Vector ◽  
Multilayer Perceptrons ◽  
Learning Problem ◽  
Stock Trading ◽  
Expected Performance ◽  
Vector Machines ◽  
Price Trends ◽  
Input Variables

There have been many machine learning-based studies to forecast stock price trends. These studies attempted to extract input features mostly from the price information with little focus on the trading volume information. In addition, modeling parameters to specify a learning problem have not been intensively investigated. We herein develop an improved method by handling those limitations. Specifically, we generated input variables by considering both price and volume information with even weight. We also defined three modeling parameters: the input and the target window sizes and the profit threshold. These specify the input and target variables, between which the underlying functions are learned by multilayer perceptrons and support vector machines. We tested our approach over six stocks and 15 years and compared with the expected performance over all considered parameter specifications. Our approach dramatically improved the prediction accuracy over the expected performance. In addition, our approach was shown to be stably more profitable than both the expected performance and the buy-and-hold strategy. On the other hand, the performance was degraded when the input variables generated from the trading volume were excluded from learning. All these results validate the importance of the volume and the modeling parameters in stock trading prediction.

scholarly journals Multiscale Supervised Classification of Point Clouds with Urban and Forest Applications

Sensors ◽  
2019 ◽  
Vol 19 (20) ◽  
pp. 4523 ◽  
Author(s):  
Carlos Cabo ◽  
Celestino Ordóñez ◽  
Fernando Sáchez-Lasheras ◽  
Javier Roca-Pardiñas ◽  
and Javier de Cos-Juez
Keyword(s):  
Random Forest ◽  
Laser Scanning ◽  
Supervised Classification ◽  
Computing Time ◽  
Principal Component ◽  
Point Clouds ◽  
Support Vector ◽  
Linear Discriminant ◽  
Vector Machines ◽  
Input Variables

We analyze the utility of multiscale supervised classification algorithms for object detection and extraction from laser scanning or photogrammetric point clouds. Only the geometric information (the point coordinates) was considered, thus making the method independent of the systems used to collect the data. A maximum of five features (input variables) was used, four of them related to the eigenvalues obtained from a principal component analysis (PCA). PCA was carried out at six scales, defined by the diameter of a sphere around each observation. Four multiclass supervised classification models were tested (linear discriminant analysis, logistic regression, support vector machines, and random forest) in two different scenarios, urban and forest, formed by artificial and natural objects, respectively. The results obtained were accurate (overall accuracy over 80% for the urban dataset, and over 93% for the forest dataset), in the range of the best results found in the literature, regardless of the classification method. For both datasets, the random forest algorithm provided the best solution/results when discrimination capacity, computing time, and the ability to estimate the relative importance of each variable are considered together.

scholarly journals Classification of Incidental Carcinoma of the Prostate Using Learning Vector Quantization and Support Vector Machines

2004 ◽  
Vol 26 (1-2) ◽  
pp. 45-55
Author(s):  
Torsten Mattfeldt ◽  
Danilo Trijic ◽  
Hans‐Werner Gottfried ◽  
Hans A. Kestler
Keyword(s):  
Support Vector Machines ◽  
Gleason Score ◽  
Vector Quantization ◽  
Volume Fraction ◽  
Learning Vector Quantization ◽  
Support Vector ◽  
Stepwise Logistic Regression ◽  
P53 Overexpression ◽  
Vector Machines ◽  
Input Variables

The subclassification of incidental prostatic carcinoma into the categories T1a and T1b is of major prognostic and therapeutic relevance. In this paper an attempt was made to find out which properties mainly predispose to these two tumor categories, and whether it is possible to predict the category from a battery of clinical and histopathological variables using newer methods of multivariate data analysis. The incidental prostatic carcinomas of the decade 1990–99 diagnosed at our department were reexamined. Besides acquisition of routine clinical and pathological data, the tumours were scored by immunohistochemistry for proliferative activity and p53‐overexpression. Tumour vascularization (angiogenesis) and epithelial texture were investigated by quantitative stereology. Learning vector quantization (LVQ) and support vector machines (SVM) were used for the purpose of prediction of tumour category from a set of 10 input variables (age, Gleason score, preoperative PSA value, immunohistochemical scores for proliferation and p53‐overexpression, 3 stereological parameters of angiogenesis, 2 stereological parameters of epithelial texture). In a stepwise logistic regression analysis with the tumour categories T1a and T1b as dependent variables, only the Gleason score and the volume fraction of epithelial cells proved to be significant as independent predictor variables of the tumour category. Using LVQ and SVM with the information from all 10 input variables, more than 80 of the cases could be correctly predicted as T1a or T1b category with specificity, sensitivity, negative and positive predictive value from 74–92%. Using only the two significant input variables Gleason score and epithelial volume fraction, the accuracy of prediction was not worse. Thus, descriptive and quantitative texture parameters of tumour cells are of major importance for the extent of propagation in the prostate gland in incidental prostatic adenocarcinomas. Classical statistical tools and neuronal approaches led to consistent conclusions.

The impact of different parameter sets on the classification of asteroid types

2021 ◽  
Author(s):  
Hanna Klimczak ◽  
Wojciech Kotłowski ◽  
Dagmara Oszkiewicz ◽  
Francesca DeMeo ◽  
Agnieszka Kryszczyńska ◽  
...  
Keyword(s):  
Gradient Boosting ◽  
Support Vector ◽  
Multilayer Perceptrons ◽  
Machine Learning Methods ◽  
Vector Machines ◽  
Science Centre ◽  
The Difference ◽  
The Impact

<p>The aim of the project is the classification of asteroids according to the most commonly used asteroid taxonomy (Bus-Demeo et al. 2009) with the use of various machine learning methods like Logistic Regression, Naive Bayes, Support Vector Machines, Gradient Boosting and Multilayer Perceptrons. Different parameter sets are used for classification in order to compare the quality of prediction with limited amount of data, namely the difference in performance between using the 0.45mu to 2.45mu spectral range and multiple spectral features, as well as performing the Prinicpal Component Analysis to reduce the dimensions of the spectral data.</p> <p> </p> <p>This work has been supported by grant No. 2017/25/B/ST9/00740 from the National Science Centre, Poland.</p>

MAPPED LEAST SQUARES SUPPORT VECTOR MACHINE REGRESSION

2005 ◽  
Vol 19 (03) ◽  
pp. 459-475 ◽  
Author(s):  
SHENG ZHENG ◽  
YUQIU SUN ◽  
JINWEN TIAN ◽  
JAIN LIU
Keyword(s):  
Least Squares ◽  
Matrix Multiplication ◽  
Learning Problems ◽  
Support Vector ◽  
Training Algorithms ◽  
Learning Problem ◽  
Multiplication Operation ◽  
Vector Machines ◽  
The Matrix ◽  
Set Of Points

This paper describes a novel version of regression SVM (Support Vector Machines) that is based on the least-squares error. We show that the solution of this optimization problem can be obtained easily once the inverse of a certain matrix is computed. This matrix, however, depends only on the input vectors, but not on the labels. Thus, if many learning problems with the same set of input vectors but different sets of labels have to be solved, it makes sense to compute the inverse of the matrix just once and then use it for computing all subsequent models. The computational complexity to train an regression SVM can be reduced to O (N2), just a matrix multiplication operation, and thus probably faster than known SVM training algorithms that have O (N2) work with loops. We describe applications from image processing, where the input points are usually of the form {(x0 + dx, y0 + dy) : |dx| < m, |dy| < n} and all such set of points can be translated to the same set {(dx, dy) : |dx| < m, |dy| < n} by subtracting (x0, y0) from all the vectors. The experimental results demonstrate that the proposed approach is faster than those processing each learning problem separately.

scholarly journals An Equivalence Between Sparse Approximation and Support Vector Machines

1998 ◽  
Vol 10 (6) ◽  
pp. 1455-1480 ◽  
Author(s):  
Federico Girosi
Keyword(s):  
Support Vector Machines ◽  
Sparse Approximation ◽  
Basis Functions ◽  
Approximation Schemes ◽  
Support Vector ◽  
Multilayer Perceptrons ◽  
Basis Pursuit ◽  
Large Set ◽  
Regularization Theory ◽  
Vector Machines

This article shows a relationship between two different approximation techniques: the support vector machines (SVM), proposed by V. Vapnik (1995) and a sparse approximation scheme that resembles the basis pursuit denoising algorithm (Chen, 1995; Chen, Donoho, & Saunders, 1995). SVM is a technique that can be derived from the structural risk minimization principle (Vapnik, 1982) and can be used to estimate the parameters of several different approximation schemes, including radial basis functions, algebraic and trigonometric polynomials, B-splines, and some forms of multilayer perceptrons. Basis pursuit denoising is a sparse approximation technique in which a function is reconstructed by using a small number of basis functions chosen from a large set (the dictionary). We show that if the data are noiseless, the modified version of basis pursuit denoising proposed in this article is equivalent to SVM in the following sense: if applied to the same data set, the two techniques give the same solution, which is obtained by solving the same quadratic programming problem. In the appendix, we present a derivation of the SVM technique in the framework of regularization theory, rather than statistical learning theory, establishing a connection between SVM, sparse approximation, and regularization theory.

scholarly journals Estimation of soil types by non linear analysis of remote sensing data

2008 ◽  
Vol 15 (1) ◽  
pp. 115-126 ◽  
Author(s):  
C. Hahn ◽  
R. Gloaguen
Keyword(s):  
Remote Sensing ◽  
Soil Type ◽  
Remote Sensing Data ◽  
Soil Types ◽  
Support Vector ◽  
Sensing Data ◽  
Vector Machines ◽  
Non Linear ◽  
Input Variables ◽  
Type Classification

Abstract. The knowledge of soil type and soil texture is crucial for environmental monitoring purpose and risk assessment. Unfortunately, their mapping using classical techniques is time consuming and costly. We present here a way to estimate soil types based on limited field observations and remote sensing data. Due to the fact that the relation between the soil types and the considered attributes that were extracted from remote sensing data is expected to be non-linear, we apply Support Vector Machines (SVM) for soil type classification. Special attention is drawn to different training site distributions and the kind of input variables. We show that SVM based on carefully selected input variables proved to be an appropriate method for soil type estimation.

scholarly journals ANALISIS LIKUIDITAS SAHAM SEBELUM, SAAT DAN SESUDAH PENGUMUMAN STOCH SPLIT

2008 ◽  
Vol 8 (1) ◽  
pp. 129
Author(s):  
Agus Sucipto
Keyword(s):  
Stock Price ◽  
Trading Volume ◽  
Market Reaction ◽  
Volume Activity ◽  
Stock Trading ◽  
Stock Liquidity ◽  
Bid Ask Spread ◽  
Significant Difference ◽  
Stock Split ◽  
Before And After

<p class="Bodytext20">Stock split announcement is one of information type published by emitent that is used to know market reaction. When stock split announcement contains information, the market reacts that is shown by the changing of stock price. This study is intended to describe the effect of stock split announcement to market reaction using event study. This approach is used to identify the reaction of the market which is an activity of trading volume and bid-ask spread of stock used to know stock liquidity. The findings show that there is no significant difference between stock trading volume activity before, during and after stock split announcement. Whereas, the period of before and after the announcement, there is a significant difference of stock trading volume activity. The finding of bid-ask spread stock shows that there is a significant difference in the period of before and after stock split announcement. But there is no significant difference in the period of before and after stock split announcement.</p><p class="Bodytext20"> </p><p class="Bodytext20">Pengumuman pemecahan saham adalah salah satu jenis informasi yang diterbitkan oleh emiten yang digunakan untuk mengetahui reaksi pasar. Bila pengumuman pemecahan saham berisi informasi, pasar bereaksi yang ditunjukkan oleh perubahan harga saham. Penelitian ini bertujuan untuk mendeskripsikan efek pengumuman pemecahan saham terhadap reaksi pasar dengan menggunakan kajian peristiwa. Pendekatan ini digunakan untuk mengidentifikasi reaksi pasar yang merupakan aktivitas volume perdagangan dan pemecahan saham yang digunakan untuk mengetahui likuiditas saham. Temuan menunjukkan bahwa tidak ada perbedaan yang signifikan antara aktivitas volume perdagangan saham sebelum, selama dan setelah pengumuman pemecahan saham. Padahal, periode sebelum dan sesudah pengumuman, ada perbedaan yang signifikan dari aktivitas volume perdagangan saham. Temuan menunjukkan bahwa ada perbedaan yang signifikan pada periode sebelum dan sesudah pengumuman pemecahan saham. Namun tidak ada perbedaan yang signifikan pada periode sebelum dan sesudah pengumuman pemecahan saham.</p>

Sign in / Sign up

Export Citation Format

Share Document