arXiv subjects

Jose M. Pena

Publications and source records attributed to Jose M. Pena.

8 recordsLinked to original sources

arXiv2025

Navigating Unmeasured Confounding in Quantitative Sociology: A Sensitivity Framework

Unmeasured confounding remains a critical challenge in causal inference for the social sciences. This paper proposes a sensitivity analysis framework to systematically evaluate how unmeasured confounders influence statistical inference in sociology. Given these sensitivity analysis methods, we introduce a five-step workflow that integrates sensitivity analysis into research design rather than treating it as a post-hoc robustness check. Using the Blau and Duncan (1967) study as an empirical example, we demonstrate how different sensitivity methods provide complementary insights. By extending existing frameworks, we show how sensitivity analysis enhances causal transparency, offering a practical tool for assessing uncertainty in observational research. Our approach contributes to a more rigorous application of causal inference in sociology, bridging gaps between theory, identification strategies, and statistical modeling.

stat.ME↗

arXiv2022

Personalized Public Policy Analysis in Social Sciences using Causal-Graphical Normalizing Flows

Structural Equation/Causal Models (SEMs/SCMs) are widely used in epidemiology and social sciences to identify and analyze the average causal effect (ACE) and conditional ACE (CACE). Traditional causal effect estimation methods such as Inverse Probability Weighting (IPW) and more recently Regression-With-Residuals (RWR) are widely used - as they avoid the challenging task of identifying the SCM parameters - to estimate ACE and CACE. However, much work remains before traditional estimation methods can be used for counterfactual inference, and for the benefit of Personalized Public Policy Analysis (P$^3$A) in the social sciences. While doctors rely on personalized medicine to tailor treatments to patients in laboratory settings (relatively closed systems), P$^3$A draws inspiration from such tailoring but adapts it for open social systems. In this article, we develop a method for counterfactual inference that we name causal-Graphical Normalizing Flow (c-GNF), facilitating P$^3$A. First, we show how c-GNF captures the underlying SCM without making any assumption about functional forms. Second, we propose a novel dequantization trick to deal with discrete variables, which is a limitation of normalizing flows in general. Third, we demonstrate in experiments that c-GNF performs on-par with IPW and RWR in terms of bias and variance for estimating the ATE, when the true functional forms are known, and better when they are unknown. Fourth and most importantly, we conduct counterfactual inference with c-GNFs, demonstrating promising empirical performance. Because IPW and RWR, like other traditional methods, lack the capability of counterfactual inference, c-GNFs will likely play a major role in tailoring personalized treatment, facilitating P$^3$A, optimizing social interventions - in contrast to the current `one-size-fits-all' approach of existing methods.

cs.LG↗

arXiv2015

High-Resolution Spectroscopic Study of Extremely Metal-Poor Star Candidates from the SkyMapper Survey

The SkyMapper Southern Sky Survey is carrying out a search for the most metal-poor stars in the Galaxy. It identifies candidates by way of its unique filter set that allows for estimation of stellar atmospheric parameters. The set includes a narrow filter centered on the Ca II K 3933A line, enabling a robust estimate of stellar metallicity. Promising candidates are then confirmed with spectroscopy. We present the analysis of Magellan-MIKE high-resolution spectroscopy of 122 metal-poor stars found by SkyMapper in the first two years of commissioning observations. 41 stars have [Fe/H] <= -3.0. Nine have [Fe/H] <= -3.5, with three at [Fe/H] ~ -4. A 1D LTE abundance analysis of the elements Li, C, Na, Mg, Al, Si, Ca, Sc, Ti, Cr, Mn, Co, Ni, Zn, Sr, Ba and Eu shows these stars have [X/Fe] ratios typical of other halo stars. One star with low [X/Fe] values appears to be "Fe-enhanced," while another star has an extremely large [Sr/Ba] ratio: >2. Only one other star is known to have a comparable value. Seven stars are "CEMP-no" stars ([C/Fe] > 0.7, [Ba/Fe] < 0). 21 stars exhibit mild r-process element enhancements (0.3 <=[Eu/Fe] < 1.0), while four stars have [Eu/Fe] >= 1.0. These results demonstrate the ability to identify extremely metal-poor stars from SkyMapper photometry, pointing to increased sample sizes and a better characterization of the metal-poor tail of the halo metallicity distribution function in the future.

astro-ph.SR↗

arXiv2015

Metal-Poor Stars Observed with the Magellan Telescope. III. New Extremely and Ultra Metal-Poor Stars from SDSS/SEGUE and Insights on the Formation of Ultra Metal-Poor Stars

We report the discovery of one extremely metal-poor (EMP; [Fe/H]<-3) and one ultra metal-poor (UMP; [Fe/H]<-4) star selected from the SDSS/SEGUE survey. These stars were identified as EMP candidates based on their medium-resolution (R~2,000) spectra, and were followed-up with high-resolution (R~35,000) spectroscopy with the Magellan-Clay Telescope. Their derived chemical abundances exhibit good agreement with those of stars with similar metallicities. We also provide new insights on the formation of the UMP stars, based on comparison with a new set of theoretical models of supernovae nucleosynthesis. The models were matched with 20 UMP stars found in the literature, together with one of the program stars (SDSS J1204+1201), with [Fe/H]=-4.34. From fitting their abundances, we find that the supernovae progenitors, for stars where carbon and nitrogen are measured, had masses ranging from 20.5 M_sun to 28 M_sun and explosion energies from 0.3 to 0.9x10^51 erg. These results are highly sensitive to the carbon and nitrogen abundance determinations, which is one of the main drivers for future high-resolution follow-up of UMP candidates. In addition, we are able to reproduce the different CNO abundance patterns found in UMP stars with a single progenitor type, by varying its mass and explosion energy.

astro-ph.SR↗

arXiv2013

Combinatorial Optimization by Learning and Simulation of Bayesian Networks

This paper shows how the Bayesian network paradigm can be used in order to solve combinatorial optimization problems. To do it some methods of structure learning from data and simulation of Bayesian networks are inserted inside Estimation of Distribution Algorithms (EDA). EDA are a new tool for evolutionary computation in which populations of individuals are created by estimation and simulation of the joint probability distribution of the selected individuals. We propose new approaches to EDA for combinatorial optimization based on the theory of probabilistic graphical models. Experimental results are also presented.

cs.AI↗

arXiv2012

On Local Optima in Learning Bayesian Networks

This paper proposes and evaluates the k-greedy equivalence search algorithm (KES) for learning Bayesian networks (BNs) from complete data. The main characteristic of KES is that it allows a trade-off between greediness and randomness, thus exploring different good local optima. When greediness is set at maximum, KES corresponds to the greedy equivalence search algorithm (GES). When greediness is kept at minimum, we prove that under mild assumptions KES asymptotically returns any inclusion optimal BN with nonzero probability. Experimental results for both synthetic and real data are reported showing that KES often finds a better local optima than GES. Moreover, we use KES to experimentally confirm that the number of different local optima is often huge.

cs.LG↗

arXiv2012

Identifying the Relevant Nodes Without Learning the Model

We propose a method to identify all the nodes that are relevant to compute all the conditional probability distributions for a given set of nodes. Our method is simple, effcient, consistent, and does not require learning a Bayesian network first. Therefore, our method can be applied to high-dimensional databases, e.g. gene expression databases.

cs.LG↗

arXiv2012

Reading Dependencies from Polytree-Like Bayesian Networks

We present a graphical criterion for reading dependencies from the minimal directed independence map G of a graphoid p when G is a polytree and p satisfies composition and weak transitivity. We prove that the criterion is sound and complete. We argue that assuming composition and weak transitivity is not too restrictive.

cs.AI↗