Large Set Of Predictors Research Articles

Updating mortality risk estimation in intensive care units from high-dimensional electronic health records with incomplete data

BackgroundThe risk of mortality in intensive care units (ICUs) is currently addressed by the implementation of scores using admission data. Their performances are satisfactory when complications occur early after admission; however, they may become irrelevant in the case of long hospital stays. In this study, we developed predictive models of short-term mortality in the ICU from longitudinal data.MethodsUsing data collected throughout patients’ stays of at least 48 h from the MIMIC-III database, several statistical learning approaches were compared, including deep neural networks and penalized regression. Missing data were handled using complete-case analysis or multiple imputation.ResultsComplete-case analyses from 19 predictors showed good discrimination (AUC > 0.77 for several approaches) to predict death between 12 and 24 h onward, yet excluded 75% of patients from the initial target cohort, as data was missing for some of the predictors. Multiple imputation allowed us to include 70 predictors and keep 95% of patients, with similar performances.ConclusionThis proof-of-concept study supports that automated analysis of electronic health records can be of great interest throughout patients’ stays as a surveillance tool. Although this framework relies on a large set of predictors, it is robust to data imputation and may be effective early after admission, when data are still scarce.

Forecasting stock volatility with a large set of predictors: A new forecast combination method

AbstractThis paper investigates how to improve the prediction accuracy of stock realized volatility using a large set of predictors. Exploiting normalized positive adjusted and significant statistic of predictor obtained from the in‐sample result as weight, we develop two simple and effective forecast combination methods. Using an array of 86 equity, bond, forex, futures, behavior, macroeconomic, and uncertainty predictors, we find that the proposed methodologies significantly improve stock realized volatility out‐of‐sample prediction performance relative to several extant forecast combinations. This result is robust for different individual forecast models, different dependent variables, and different out‐of‐sample periods. Furthermore, we explain that out‐of‐sample predictability varies significantly with changes in the number of predictors. And the existence of a strongly powerful volatility predictor affects this change.

Conspiracist and paranormal beliefs: A typology of non-reductive ideation

Non-reductive ideation is characterized by a tendency to attribute causality to non-physical powers and mysterious forms of intelligent agency such as conspiracist ideation, belief in specific conspiracy theories, and belief in the paranormal. Scholars have identified numerous individual correlates of non-reductive ideation but do not examine shared patterns across multiple predictors to determine subtypes of individuals with different psychological profiles. We address this gap by considering a large set of predictors in a diverse cross-section of the US public (N=792) to uncover latent subtypes of individuals with varying tendencies toward non-reductive ideation. Schizotypal and neurotic tendencies were the strongest predictors of non-reductive ideation, while sociopolitical identity (e.g., political ideology) contributed little explanatory power. We find five distinct latent classes distinguished by: schizotypal and paranoid ideation, alienated skepticism toward people/society, and a negative sense of self. We discuss these results in light of previous findings and suggest directions for future research.

Determinants of cryptocurrency returns: A LASSO quantile regression approach

We consider a relatively large set of predictors and investigate the determinants of cryptocurrency returns at different quantiles. Our analysis exclusively focuses on the highly volatile period of COVID-19. The innovation in the paper stems from the fact that we employ the LASSO penalty in a quantile regression framework to select informative variables. We find that US government bond indices and small company stock returns, a new predictor introduce in this study, significantly impact the tail behavior of the cryptocurrency returns.

Forecasting crude oil volatility with exogenous predictors: As good as it GETS?

This paper aims to investigate the usefulness of exogenous predictors to forecast crude oil volatility. We use the recent expansion of the general-to-specific (GETS) procedure to conditionally heteroskedastic models to estimate a parsimonious predictive model of crude oil volatility from a large set of predictors. Our results show that the GETS algorithm achieves good predictive accuracy compared to its competitors at the 1-day horizon. However, this accuracy deteriorates for more distant forecast horizons. We argue that it may be due to the fact that the GETS procedure is based on tests that are key in assessing explanatory power as opposed to reducing expected prediction error. Among its competitors, DMA achieves good predictive power in almost all situations. Still, our analysis provides interesting insights on the variables best suited to forecast crude oil volatility. In particular, forecasters might benefit from better exploiting the predictive content of exchange rates.

Predicting How Well Adolescents Get Along with Peers and Teachers: A Machine Learning Approach.

How well adolescents get along with others such as peers and teachers is an important aspect of adolescent development. Current research on adolescent relationship with peers and teachers is limited by classical methods that lack explicit test of predictive performance and cannot efficiently discover complex associations with potential non-linearity and higher-order interactions among a large set of predictors. Here, a transparently reported machine learning approach is utilized to overcome these limitations in concurrently predicting how well adolescents perceive themselves to get along with peers and teachers. The predictors were 99 items from four instruments examining internalizing and externalizing psychopathology, sensation-seeking, peer pressure, and parent-child conflict. The sample consisted of 3232 adolescents (M = 14.0 years, SD = 1.0 year, 49% female). Nonlinear machine learning classifiers predicted with high performance adolescent relationship with peers and teachers unlike classical methods. Using model explainability analyses at the item level, results identified influential predictors related to somatic complaints and attention problems that interacted in nonlinear ways with internalizing behaviors. In many cases, these intrapersonal predictors outcompeted in predictive power many interpersonal predictors. Overall, the results suggest the need to cast a much wider net of variables for understanding and predicting adolescent relationships, and highlight the power of a data-driven machine learning approach with implications on a predictive science of adolescence research.

Quantile Factor Models

Quantile factor models (QFM) represent a new class of factor models for high‐dimensional panel data. Unlike approximate factor models (AFM), which only extract mean factors, QFM also allow unobserved factors to shift other relevant parts of the distributions of observables. We propose a quantile regression approach, labeled Quantile Factor Analysis (QFA), to consistently estimate all the quantile‐dependent factors and loadings. Their asymptotic distributions are established using a kernel‐smoothed version of the QFA estimators. Two consistent model selection criteria, based on information criteria and rank minimization, are developed to determine the number of factors at each quantile. QFA estimation remains valid even when the idiosyncratic errors exhibit heavy‐tailed distributions. An empirical application illustrates the usefulness of QFA by highlighting the role of extra factors in the forecasts of U.S. GDP growth and inflation rates using a large set of predictors.

Forecasting Stock Volatility with a Large Set of Predictors: A New Forecast Combination Method

Forecasting Stock Volatility with a Large Set of Predictors: A New Forecast Combination Method

Equity premium prediction and optimal portfolio decision with Bagging

We propose using the statistical method of Bagging to forecast the equity premium out-of-sample for multivariate regression models. Bagging allows for the flexible and efficient extraction of valuable informational content from a large set of predictors, leading to statistically and economically significant gains relative to not only the historical mean, but also other soft-threshold methods such as forecast combinations and shrinkage estimators in our empirical results. Furthermore, we find that the source of economic gains for Bagging primarily comes from the fact that it encourages the investor to actively manage portfolio by flexibly utilizing short selling or leveraging to better time the market following correctly prognosticated trends. However, other strategies such as forecast combinations keep the equity shares nearly fixed regardless of the predicted market prospect.

Macroeconomic Forecasting with Factor-Augmented Adjusted Band Regression

Previous findings indicate that the inclusion of dynamic factors obtained from a large set of predictors can improve macroeconomic forecasts. In this paper, we explore three possible further developments: (i) using automatic criteria for choosing those factors which have the greatest predictive power; (ii) using only a small subset of preselected predictors for the calculation of the factors; and (iii) utilizing frequency-domain information for the estimation of the factor models. Reanalyzing a standard macroeconomic dataset of 143 U.S. time series and using the major measures of economic activity as dependent variables, we find that (i) is not helpful, whereas focusing on the low-frequency components of the factors and disregarding the high-frequency components can actually improve the forecasting performance for some variables. In the case of the gross domestic product, a combination of (ii) and (iii) yields the best results.

Forecasting crude oil prices with a large set of predictors: Can LASSO select powerful predictors?

Abstract In this paper, we use two prevailing shrinkage methods, the lasso and elastic net, to predict oil price returns with a large set of predictors. The out-of-sample results indicate that the lasso and elastic net models outperform a host of widely used competing models in terms of out-of-sample R-square and success ratio. In an asset allocation exercise, a mean–variance investor obtains positive and sizeable economic gains based on the return forecasts of the lasso and elastic net methods relative to both the benchmark forecasts and competing forecasts. We further investigate the source of predictability from a variable selection perspective. The lasso and elastic net methods are found to select powerful predictors and the ones that can provide complementary information. The OLS regression models based on the selected predictors also exhibit better out-of-sample performances than the competing models. In addition, our results are robust to various settings.

Assessment of factors driving high fire severity potential and classification in a Mediterranean pine ecosystem

Fire severity is an increasingly critical issue for forest managers for estimating fire impacts. Estimating high fire severity potential and accurate classification between fire severity levels are essential for integrated fire management planning in fire prone Mediterranean pine ecosystems. This study attempts to determine the role of topography, pre-fire forest stand structure, fuel complex characteristics and fire behavior parameters on high fire severity potential and classification based on a large fire event occurred in Thasos, Greece. Within this framework, the Random Forest (RF) classification algorithm was used to model the relationship between a large set of predictors and fire severity as expressed by the differenced Normalized Burn Ratio (dNBR) spectral index, inferred from differenced pre- and post-fire Landsat 8 Operational Land Imager (OLI) at 30-m resolution. Results from the RF classifier algorithm showed that high fire severity potential and classification between fire severity levels mainly depended on topography variables and fuel complex characteristics. Assessing of factors which drive a fire to turn into high severe fire and classification into fire severity levels will substantially help land and forest managers to increase fire prevention and develop of concrete actions for successful post fire management at landscape level.

Forecasting the prices of crude oil: An iterated combination approach

In this paper, we employ an iterated combination approach to examine oil price predictability with a large set of predictors, including 18 macroeconomic variables and 18 technical indicators. The empirical results show that iterated combination approach outperforms the standard combination approach for both in- and out-of-sample. Specifically, the iterated combination forecasts always yield significantly larger out-of-sample R2 values and higher success ratios than the corresponding standard combination forecasts. Furthermore, we document that the results are robust to various settings, including alternative proxies of crude oil prices, three predictor sets, different forecasting windows, and various standard combination approaches. From an asset allocation perspective, we measure the economic value of the iterated combination approaches, where the leverage of oil futures trading is considered. The results suggest that the more accurate forecasts of the iterated combination approaches can generate substantially larger certainty equivalent return (CER) gains for a mean-variance investor in practice.

Sparse Bayesian Variable Selection in Probit Model for Forecasting U.S. Recessions Using a Large Set of Predictors

In this paper, a large set of macroeconomic and financial predictors is used to forecast U.S. recession periods. We propose a sparse Bayesian variable selection in probit model for predicting U.S. recessions. The correlation prior is assigned for the binary vector to distinguish models with the same size, and the sparse prior is specified for the coefficient parameters for the purpose of predicting accurately using fewer parameters. In terms of the quadratic probability score and the log probability score, we demonstrate that the proposed method performs better than other three methods.

Forecasting US Recessions with a Large Set of Predictors

In this paper, I use a large set of macroeconomic and financial predictors to forecast US recession periods. I adopt Bayesian methodology with shrinkage in the parameters of the probit model for the binary time series tracking the state of the economy. The in‐sample and out‐of‐sample results show that utilizing a large cross‐section of indicators yields superior US recession forecasts in comparison to a number of parsimonious benchmark models. Moreover, the data‐rich probit model gives similar accuracy to the factor‐based model for the 1‐month‐ahead forecasts, while it provides superior performance for 1‐year‐ahead predictions. Finally, in a pseudo‐real‐time application for the Great Recession, I find that the large probit model with shrinkage is able to pick up the recession signals in a timely fashion and does well in comparison to the more parsimonious specification and to nonparametric alternatives. Copyright © 2016 John Wiley & Sons, Ltd.

Forecasting macroeconomic variables in data-rich environments

We show that high-dimensional models produce, on average, smaller forecasting errors for macroeconomic variables when we consider a large set of predictors. Our results showed that a good selection of the adaptive LASSO hyperparameters also reduces forecast errors.

MT-HESS: an efficient Bayesian approach for simultaneous association detection in OMICS datasets, with application to eQTL mapping in multiple tissues.

Motivation: Analysing the joint association between a large set of responses and predictors is a fundamental statistical task in integrative genomics, exemplified by numerous expression Quantitative Trait Loci (eQTL) studies. Of particular interest are the so-called ‘hotspots’, important genetic variants that regulate the expression of many genes. Recently, attention has focussed on whether eQTLs are common to several tissues, cell-types or, more generally, conditions or whether they are specific to a particular condition.Results: We have implemented MT-HESS, a Bayesian hierarchical model that analyses the association between a large set of predictors, e.g. SNPs, and many responses, e.g. gene expression, in multiple tissues, cells or conditions. Our Bayesian sparse regression algorithm goes beyond ‘one-at-a-time’ association tests between SNPs and responses and uses a fully multivariate model search across all linear combinations of SNPs, coupled with a model of the correlation between condition/tissue-specific responses. In addition, we use a hierarchical structure to leverage shared information across different genes, thus improving the detection of hotspots. We show the increase of power resulting from our new approach in an extensive simulation study. Our analysis of two case studies highlights new hotspots that would remain undetected by standard approaches and shows how greater prediction power can be achieved when several tissues are jointly considered.Availability and implementation: C source code and documentation including compilation instructions are available under GNU licence at http://www.mrc-bsu.cam.ac.uk/software/.Contact: sylvia.richardson@mrc-bsu.cam.ac.uk or lb664@cam.ac.ukSupplementary information: Supplementary data are available at Bioinformatics online.

Forecasting with Sufficient Dimension Reductions

Factor models have been successfully employed in summarizing large datasets with few underlying latent factors and in building time series forecasting models for economic variables. When the objective is to forecast a target variable y with a large set of predictors x, the construction of the summary of the xs should be driven by how informative on y it is. Most existing methods first reduce the predictors and then forecast y in independent phases of the modeling process. In this paper we present an alternative and potentially more attractive alternative: summarizing x as it relates to y, so that all the information in the conditional distribution of y|x is preserved. These y-targeted reductions of the predictors are obtained using Sufficient Dimension Reduction techniques. We show in simulations and real data analysis that forecasting models based on sufficient reductions have the potential of significantly improved performance.

Dynamic Model Averaging and CPI Inflation Forecasts: A Comparison between the Euro Area and the United States

AbstractThe paper forecasts consumer price inflation in the euro area (EA) and in the USA between 1980:Q1 and 2012:Q4 based on a large set of predictors, with dynamic model averaging (DMA) and dynamic model selection (DMS). DMA/DMS allows not solely for coefficients to change over time, but also for changes in the entire forecasting model over time. DMA/DMS provides on average the best inflation forecasts with regard to alternative approaches (such as the random walk). DMS outperforms DMA. These results are robust for different sample periods and for various forecast horizons. The paper highlights common features between the USA and the EA. First, two groups of predictors forecast inflation: temporary fundamentals that have a frequent impact on inflation but only for short time periods; and persistent fundamentals whose switches are less frequent over time. Second, the importance of some variables (particularly international food commodity prices, house prices and oil prices) as predictors for consumer price index inflation increases when such variables experience large shocks. The paper also shows that significant differences prevail in the forecasting models between the USA and the EA. Such differences can be explained by the structure of these respective economies. Copyright © 2015 John Wiley & Sons, Ltd.

Forecasting with Sufficient Dimension Reductions

Factor models have been successfully employed in summarizing large datasets with few underlying latent factors and in building time series forecasting models for economic variables. When the objective is to forecast a target variable y with a large set of predictors x, the construction of the summary of the xs should be driven by how informative on y it is. Most existing methods first reduce the predictors and then forecast y in independent phases of the modeling process. In this paper we present an alternative and potentially more attractive alternative: summarizing x as it relates to y, so that all the information in the conditional distribution of yjx is preserved. These y-targeted reductions of the predictors are obtained using Sufficient Dimension Reduction techniques. We show in simulations and real data analysis that forecasting models based on sufficient reductions have the potential of significantly improved performance.