scholarly journals High‐dimensional macroeconomic forecasting and variable selection via penalized regression

2019 ◽  
Vol 22 (1) ◽  
pp. 34-56 ◽  
Author(s):  
Yoshimasa Uematsu ◽  
Shinya Tanaka
Keyword(s):  
New York ◽  
Variable Selection ◽  
Stock Price ◽  
Stock Exchange ◽  
Penalized Regression ◽  
Mixed Data ◽  
High Dimensional ◽  
Data Sampling ◽  
Data Set ◽  
Macroeconomic Forecasting

Summary This study examines high-dimensional forecasting and variable selection via folded-concave penalized regressions. The penalized regression approach leads to sparse estimates of the regression coefficients and allows the dimensionality of the model to be much larger than the sample size. First, we discuss the theoretical aspects of a penalized regression in a time series setting. Specifically, we show the oracle inequality with ultra-high-dimensional time-dependent regressors. Then we show the validity of the penalized regression using two empirical applications. First, we forecast quarterly US gross domestic product data using a high-dimensional monthly data set and the mixed data sampling (MIDAS) framework with penalization. Second, we examine how well the penalized regression screens a hidden portfolio based on a large New York Stock Exchange stock price data set. Both applications show that a penalized regression provides remarkable results in terms of forecasting performance and variable selection.

Pengaruh Reputasi Underwriter, Umur Perusahaan, dan Persentase Penawaran Saham 53 kepada Publik terhadap Underpricing

2016 ◽  
Vol 8 (1) ◽  
pp. 53-74
Author(s):  
Maria Jeanne ◽  
Chermian Eforis
Keyword(s):  
Initial Public Offering ◽  
Stock Price ◽  
Stock Exchange ◽  
Secondary Data ◽  
Secondary Market ◽  
Financial Statement ◽  
Data Set ◽  
Multiple Regression Method ◽  
Underwriter Reputation ◽  
Public Offering

The objective of this research is to obtain empirical evidence about the effect of underwriter reputation, company age, and the percentage of share’s offering to public toward underpricing. Underpricing is a phenomenon in which the current stock price initial public offering (IPO) was lower than the closing price of shares in the secondary market during the first day. Sample in this research was selected by using purposive sampling method and the secondary data used in this research was analyzed by using multiple regression method. The samples in this research were 72 companies conducting initial public offering (IPO) at the Indonesian Stock Exchange in the period January 2010 - December 2014; perform initial offering of shares; suffered underpricing; has a complete data set forth in the company's prospectus, IDX monthly statistics, financial statement and stock price site (e-bursa); and use Rupiah currency. Results of this research were (1) underwriter reputation significantly effect on underpricing; (2) company age do not effect on underpricing; and (3) the percentage of share’s offering to public do not effect on undepricing. Keywords: company age, the percentage of share’s offering to public, underpricing, underwriter reputation.

scholarly journals Forecasting Changes in Stock Prices on the Basis of Patterns Identified with the Use of Data Classification Methods

2014 ◽  
Vol 14 (1) ◽  
pp. 7-21
Author(s):  
Jacek Szanduła
Keyword(s):  
New York ◽  
Stock Prices ◽  
Stock Price ◽  
Stock Exchange ◽  
Data Classification ◽  
Classification Methods ◽  
Financial Instrument ◽  
Partitioning Around Medoids ◽  
Use Of Data ◽  
Short And Long Term

Abstract The paper develops the concept of harnessing data classification methods to recognize patterns in stock prices. The author defines a formation as a pattern vector describing the financial instrument. Elements of such a vector can be related to the stock price as well as sales volume and other characteristics of the financial instrument. The study uses data concerning selected companies listed on the stock exchange in New York. It takes into account a number of variables that describe the behavior of prices and volume, both in the short and long term. Partitioning around medoids method has been used for data classification (for pattern recognition). An evaluation of the possibility of using certain formations for practical purposes has also been presented.

scholarly journals The Market Reaction Around Ex-Dates of Stock Splits Before and After Decimalization

2005 ◽  
Vol 08 (02) ◽  
pp. 201-216 ◽  
Author(s):  
Robin K. Chou ◽  
Wan-Chen Lee ◽  
Sheng-Syan Chen
Keyword(s):  
New York ◽  
Stock Price ◽  
Trading Volume ◽  
Stock Exchange ◽  
New York Stock Exchange ◽  
Short Term ◽  
Tick Size ◽  
York Stock Exchange ◽  
Before And After ◽  
Abnormal Trading Volume

This paper examines the stock price behavior around the ex-split dates both before and after the decimalization on the New York Stock Exchange (NYSE). We find that the abnormal ex-split day returns decrease and the abnormal trading volume increases in the 1/16th and decimal pricing eras, relative to the 1/8th pricing era. These findings are consistent with the microstructure-based explanations for the ex-day price movements. Our study also supports the hypothesis that short-term traders perform arbitrage activities during the ex-split dates when transaction costs become lower after the tick size is reduced.

scholarly journals Time for Business: Business Ethics, Sustainability, and Giorgio Agamben’s ‘Messianic Time’

2016 ◽  
Vol 2 (3) ◽  
pp. 39-51
Author(s):  
Jeremy Kidwell
Keyword(s):  
New York ◽  
Stock Price ◽  
Stock Exchange ◽  
Recent Attempt ◽  
New York Stock Exchange ◽  
York Stock Exchange ◽  
Modern Business ◽  
Future Consequences ◽  
Economic Future ◽  
Hourly Basis

Contemporary business continues to intensify its radical relation to time. The New York Stock Exchange recently announced that in pursuing (as traders call it) the ‘race to zero’ they will begin using laser technology originally developed for military communications to send information about trades nearly at the speed of light. This is just one example of short-term temporal rhythms embedded in the practices of contemporary firms which watch their stock price on an hourly basis, report their earnings quarterly, and dissolve future consequences and costs through discounting procedures. There is reason to believe that these radical conceptions of time and its passing impair the ability of businesses to function in a morally coherent manner. In the spirit of other recent critiques of modern temporality such as David Couzen Hoys The Time of Our Lives, in this paper, I present a critique of the temporality of modern business. In response, I assess the recent attempt to provide an alternative account of temporality using theological concepts by Giorgio Agamben. I argue that Agamben’s more integrative account of messianic time provides a richer ambitemporal account which might provide a viable temporality for a new sustainable economic future.

scholarly journals A mixture of local and quadratic approximation variable selection algorithm in nonconcave penalized regression

2013 ◽  
Vol Volume 16, 2012 ◽  
Author(s):  
Assi N'GUESSAN ◽  
Ibrahim Sidi Zakari ◽  
Assi Mkhadri
Keyword(s):  
Variable Selection ◽  
Penalized Likelihood ◽  
Real Data ◽  
Penalized Regression ◽  
Penalty Functions ◽  
Selection Algorithm ◽  
Data Set ◽  
International Audience ◽  
One Step ◽  
Nonconvex Penalty

International audience We consider the problem of variable selection via penalized likelihood using nonconvex penalty functions. To maximize the non-differentiable and nonconcave objective function, an algorithm based on local linear approximation and which adopts a naturally sparse representation was recently proposed. However, although it has promising theoretical properties, it inherits some drawbacks of Lasso in high dimensional setting. To overcome these drawbacks, we propose an algorithm (MLLQA) for maximizing the penalized likelihood for a large class of nonconvex penalty functions. The convergence property of MLLQA and oracle property of one-step MLLQA estimator are established. Some simulations and application to a real data set are also presented.

Performance and development of a thin stock market: the Stockholm Stock Exchange 1912–2017

2020 ◽  
pp. 1-19
Author(s):  
Kristian Rydqvist ◽  
Rong Guo
Keyword(s):  
Stock Market ◽  
Stock Returns ◽  
Stock Prices ◽  
Stock Price ◽  
Stock Exchange ◽  
Stock Selection ◽  
Data Set ◽  
Transaction Prices ◽  
Limit Orders ◽  
A Value

We estimate historical stock returns for Swedish listed companies in a newly constructed data set of daily stock prices that spans more than 100 years. Stock returns exhibit all the familiar characteristics. The growth of the public sector depressed the stock market, and the process of globalization revitalized it. Banks played an important role in the early development of the stock market. There was little trading in the past, and we examine the effects on return measurement from missing data. Stock selection and the replacement of missing transaction prices through search back procedures or limit orders make little difference to a value-weighted stock price index, while ignoring the price effects of capital operations makes a big difference.

Variable selection for ultra-high dimensional quantile regression with missing data and measurement error

2020 ◽  
pp. 096228022094153
Author(s):  
Yongxin Bai ◽  
Maozai Tian ◽  
Man-Lai Tang ◽  
Wing-Yan Lee
Keyword(s):  
Missing Data ◽  
Measurement Error ◽  
Variable Selection ◽  
Quantile Regression ◽  
Measurement Errors ◽  
Estimation Method ◽  
High Dimensional ◽  
Data Set ◽  
Cancer Data ◽  
Selection For

In this paper, we consider variable selection for ultra-high dimensional quantile regression model with missing data and measurement errors in covariates. Specifically, we correct the bias in the loss function caused by measurement error by applying the orthogonal quantile regression approach and remove the bias caused by missing data using the inverse probability weighting. A nonconvex Atan penalized estimation method is proposed for simultaneous variable selection and estimation. With the proper choice of the regularization parameter and under some relaxed conditions, we show that the proposed estimate enjoys the oracle properties. The choice of smoothing parameters is also discussed. The performance of the proposed variable selection procedure is assessed by Monte Carlo simulation studies. We further demonstrate the proposed procedure with a breast cancer data set.

scholarly journals Pattern Of Short-Term Volatility Accentuation Within The Trading Day: An Investigation Of The U.S. And European Equity Markets

2011 ◽  
Vol 5 (2) ◽  
Author(s):  
Deniz Ozenbas
Keyword(s):  
New York ◽  
Stock Price ◽  
Price Discovery ◽  
Stock Exchange ◽  
Equity Markets ◽  
Price Volatility ◽  
Short Term ◽  
Major Index ◽  
Short Period ◽  
The Individual

Trading friction leads into accentuated stock price volatility over the short term. As such, short-term accentuated volatility can be viewed as symptomatic of a market with increased inefficiencies in the price discovery process. If price discovery is marked by price swings, runs and reversals, then short period (intra-day) volatility is heightened in that market. In this study, we use return series with various differencing intervals that are as short as half-hour and as long as two weeks to investigate the short-term volatility accentuation in five different equity markets: the Nasdaq Stock Market and the New York Stock Exchange in the US, and the London Stock Exchange, Deutsche Boerse and Euronext Paris in Europe. In all these markets, we investigate the individual stocks that make up a major index during the calendar year 2000. Variance-ratio statistics are employed to investigate the quality of these five markets. Results confirm an intra-day reverse J-shaped pattern of half-hour volatility in these markets. The evidence also suggests an accentuation of volatility during longer periods, such as 24-hour intervals. This accentuation appears to subside when we extend our differencing interval to longer periods such as one-week or two-week returns.

Sign in / Sign up

Export Citation Format

Share Document