Partial Identification Research Articles

Sharp bounds in the latent index selection model

A fundamental question underlying the literature on partial identification is: what can we learn about parameters that are relevant for policy but not necessarily point-identified by the exogenous variation we observe? This paper provides an answer in terms of sharp, analytic characterizations and bounds for an important class of policy-relevant treatment effects, consisting of marginal treatment effects and linear functionals thereof, in the latent index selection model as formalized in Vytlacil (2002). The sharp bounds use the full content of identified marginal distributions, and analytic derivations rely on the theory of stochastic orders. The proposed methods also make it possible to sharply incorporate new auxiliary assumptions on distributions into the latent index selection framework. Empirically, I apply the methods to study the effects of Medicaid on emergency room utilization in the Oregon Health Insurance Experiment, showing that the predictions from extrapolations based on a distribution assumption (rank similarity) differ substantively and consistently from existing extrapolations based on a parametric mean assumption (linearity). This underscores the value of utilizing the model’s full empirical content in combination with auxiliary assumptions.

Partial identification and inference for conditional distributions of treatment effects

SummaryThis paper considers identification and inference for the distribution of treatment effects conditional on observable covariates. Since the conditional distribution of treatment effects is not point identified without strong assumptions, we obtain bounds on the conditional distribution of treatment effects by using the Makarov bounds. We also consider the case where the treatment is endogenous and propose two stochastic dominance assumptions to tighten the bounds. We develop a nonparametric framework to estimate the bounds and establish the asymptotic theory that is uniformly valid over the support of treatment effects. An empirical example illustrates the usefulness of the methods.

A Partial Approximate Least Absolute Deviation-Based Identification Algorithm for a Multivariable Closed-Loop System with Spike Noise

Considering the stability and safety of industrial production, the open-loop system cannot meet the requirements of industrial production, so more and more attention has been paid to multivariable closed-loop system identification. This study focuses on the identification of a multivariate closed-loop system with spike noise in which the model order of the feedback channel is lower than that of the forward channel. Combining principal component analysis (PCA), which is used to eliminate the correlation between the data matrix elements, and the derivable approximate least absolute criterion function, partial approximate least absolute deviation (PALAD) for multivariable closed-loop system identification is proposed. By introducing a deterministic function to replace the absolute value in partial approximate least absolute deviation, the non-differentiable problem of the least absolute deviation function can be solved, and the identifiability of the multivariable closed-loop system in the case mentioned above is theoretically verified. Simulation experiments show the validity of the PALAD algorithm. Compared with the partial least squares (PLS) method, PALAD can effectively restrain the spike noise that follows an SαS distribution and shows stronger robustness when white and spike noises exist simultaneously.

Win/Loss Data and Consumer Switching Costs: Measuring Diversion Ratios and the Impact of Mergers*

We propose an identification strategy for diversion based on win/loss data. First, we show that win/loss data from the merging firms and market shares in two periods for all firms are sufficient to identify the diversion ratios between the merging partners. Second, we show that win/loss data from the merging firms alone are sufficient for partial identification, and we construct a lower bound that provides a good approximation to the diversion ratio when switching costs are high. We demonstrate the performance of our method with numerical simulations and with an application to the Anthem/Cigna merger.

Finite-time synchronization and topology identification of stochastic multi-layer networks with Markovian switching

The finite-time synchronization and topology identification problems of stochastic multi-layer networks with Markovian switching are investigated in this paper. Compared with previous studies, Markovian switching is considered in stochastic multi-layer networks. First, by using Kirchhoff’s matrix tree theorem in graph theory, the Lyapunov function of the coupling system is constructed indirectly from the Lyapunov function of the vertex system. In view of finite-time stability theory and stochastic analysis technique, some finite-time synchronization criteria of stochastic multi-layer networks are proposed. Moreover, the intra-layer and inter-layer topological structures are successfully identified in finite time. Then, using the pinning control technique, the unknown partial topological structures are also identified in finite time. In the end, two examples for whole and partial topology identification of two-layer small-world networks are given.

Partial identification of visible lines from W24+ ions in an electron-beam ion trap

To better understand the structure of the open shell, the spectroscopic data in the 420–600 nm range of the highly charged ions of W24+ obtained with an electron beam ion trap are presented. Detailed collisional-radiative (CR) modeling has been used to simulate the spectrum. The atomic structure data for the CR modeling were calculated using the relativistic many-body perturbation theory (RMBPT) and the large-scale relativistic configuration interaction (RCI) method implemented in the flexible atomic code. The two set theoretical calculation wavelengths for six observed lines are in reasonable agreement with the experimental observations, with average deviations of 0.61% for RMBPT and 1.75% for RCI. The relative intensities of the identified lines have also been investigated, revealing a strong correlation to electron density from = 109 to 1011 cm−3.

Treatment choice, mean square regret and partial identification

We consider a decision maker who faces a binary treatment choice when their welfare is only partially identified from data. We contribute to the literature by anchoring our finite-sample analysis on mean square regret, a decision criterion advocated by Kitagawa et al. in (2022) "Treatment Choice with Nonlinear Regret" . We find that optimal rules are always fractional, irrespective of the width of the identified set and precision of its estimate. The optimal treatment fraction is a simple logistic transformation of the commonly used t-statistic multiplied by a factor calculated by a simple constrained optimization. This treatment fraction gets closer to 0.5 as the width of the identified set becomes wider, implying the decision maker becomes more cautious against the adversarial Nature.

Recent Developments in Partial Identification

Identification strategies concern what can be learned about the value of a parameter based on the data and the model assumptions. The literature on partial identification is motivated by the fact that it is not possible to learn the exact value of the parameter for many empirically relevant cases. A typical result in the literature on partial identification is a statement about characterizing the identified set, which summarizes what can be learned about the parameter of interest given the data and model assumptions. For instance, this may mean that the value of the parameter can be learned to be necessarily within some set of values. First, the review surveys the general frameworks that have been developed for conducting a partial identification analysis. Second, the review surveys some of the more recent results on partial identification.

Causal effects of mental health on food security

Although mental health conditions are known to be associated with socioeconomic hardships, their causal effects remain largely unexplored. Using a sample of low-income families in the National Health Interview Survey (NHIS), we assess causal effects of serious mental illness (SMI) and related mental health conditions on family food security. We apply partial identification methods to account for fundamental endogeneity and measurement identification problems in a unified framework. To implement these methods, we combine a proxy measure of SMI in the NHIS with an estimate of the true rate of SMI from the Substance Abuse and Mental Health Services Administration. We also develop an innovative approach to approximate true prevalence rates when only self-reported prevalence rates are available. Applying relatively weak monotonicity assumptions on latent food security outcomes, we find that alleviating SMI would improve the food security rate by at least 9.5 percentage points, or 15 %.JEL codes: C21, I10, I38

A novel minutiae-oriented approach for partial fingerprint-based MasterPrint mitigation

Partial fingerprint identification systems recognise an individual when the sensor size has a small form factor in accepting a full fingerprint. However, the distinctive features within a partial fingerprint are significantly less. Hence, the uniqueness of a partial fingerprint cannot be assured, leading to the possibility of identifying multiple users. A MasterPrint is a partial fingerprint identifying at least 4% distinct individuals in a partial fingerprint identification system. This work addresses the MasterPrint vulnerability by proposing a novel partial fingerprint identification scheme that extracts minutiae-oriented local features from binarized and thinned partial fingerprint images over eight axes emerging from a reference minutia. It also introduces a metric to compute the similarity score between two partial fingerprint templates. The results are compared with the baseline minutiae matching (BMM) method, a modified Speeded-Up Robust Features (SURF) based approach, VeriFinger 12.1 SDK, standard NIST NBIS, and Ridge Shape Feature (RSF) scheme. The experiments employing partial fingerprint datasets cropped from standard FVC2002 DB1_A, FVC2002 DB2_A, NIST Special Databases (sd302b and sd302d), and CrossMatch VeriFinger dataset have demonstrated that the proposed method generates the lowest MasterPrints with the highest identification accuracy.

Identifying partial topology of complex dynamical networks with distributed delay

Though the discrete delay has been extensively studied in the field of network topology identification, little attention has been paid to the distributed delay. In this paper, we address the partial topology identification problem of complex dynamical networks with distributed delay. By constructing a response network having fewer nodes than the drive network, partial topology of networks with distributed delay can be identified under the widely-used linear independence condition. It is the first time that the distributed delay has been successfully addressed in the field of topology identification. Finally, we provide two numerical examples to verify the effectiveness of the theoretical results.

Allelopathic Interactions of Solanum stramonifolium Jacq. Might Be Used for Meloidogyne incognita Management

The root-knot nematode (RKN) Meloidogyne incognita infects a large host range causing enormous agricultural losses. Although the control of this pest is mainly based on the use of chemical pesticides, natural alternatives are currently being largely considered, especially the use of plant-derived molecules, as allelochemicals and plant extracts. Solanum stramonifolium is a plant of the Solanaceae family and has been studied for its potential to control RKN, either by the use aqueous extracts of its seeds, or due to its immunity to nematode penetration. Our studies demonstrate that S. stramonifolium has an efficient defense mechanism against RKN completely preventing its penetration into its root system. Additionally, the compounds exuded by this plant in the soil can affect nematode infection in nearby grown Nicotiana benthamiana (-89% biomass for treatment I respectively, and -61% and +57% of galls for the treatments and Solanum lycopersicum (-79% biomass for treatment I, and -83% of galls for the treatments I and II). Herein, external (ED, < 3.5 kDa) and internal (ID, > 3.5 kDa) dialysates were prepared from the extract of S. stramonifolium seeds. Arabidopsis thaliana Col-0 inoculated with RKN treated with the dialysates showed a reduced gall diameter in 45% and 35% for ED and ID respectively. Additionally, a significant reduction in biomass was observed in germinated seeds after treatment with ID. In addition, Arabidopsis roots showed changes in their ploidy levels when treated with ED and ID, compared to the untreated control, indicating its effect on endoreduplicating cells. Finally, a partial identification of glycoalkaloids present in the ED and root extracts may help to explain plant immunity to root-knot nematode infection. Results obtained shed light on the defense mechanisms of S. stramonifolium against M. incognita, its allelopathic potential in co-cultivated plants and its biotechnological potential of dialysates obtained from its seeds, representing a strong alternative to the management of RKN in conventional agriculture.

Characterisation of a Bacillus sp. Isolated from Soils Near Lake Maninjau Capable of Degrading Glyphosate

Glyphosate and other soil toxins can be ecologically and financially sustainable bioremediation. When it pollutes the environment, the herbicide glyphosate—used extensively for weed control—poses serious dangers to animals and their ecosystems. A rice field that had been treated with glyphosate for several years was the subject of this investigation into the bioremediation capabilities of soil bacteria. We provisionally identified the most promising isolate as Bacillus sp. strain Unand1 using partial identification methods. Under ideal circumstances, this bacterium demonstrated considerable promise for degrading glyphosate. Bacterial experiments showed that the optimal conditions for glyphosate degradation were a pH of 6.0–7.0, a glyphosate concentration of 0.5–0.6 g/L, 30 °C, and a 1% (v/v) inoculum size. At 0.5 g/L glyphosate, the bacterium showed a two-day lag phase before degrading over 90% after six days of incubation. There was a remarkable 99% and 95% suppression rate of bacterial growth, respectively, when exposed to heavy metals such as Hg(II) and Ag(I). According to this study, this bacterium has bioremediation capability for glyphosate-contaminated settings. Full characterisation and optimization of this bioremediation approach require additional research, especially using molecular identification techniques. This method has the potential to greatly reduce the negative effects of glyphosate pollution on ecosystems and agricultural sustainability.

Graph Theory-based Approach for Partial Topology Identification of Stochastic Multi-group Models With Multiple Dispersal

Graph Theory-based Approach for Partial Topology Identification of Stochastic Multi-group Models With Multiple Dispersal

Generalized synchronization-based partial topology identification of complex networks

Generalized synchronization-based partial topology identification of complex networks

Recovery of spatially varying acoustical properties via automated partial differential equation identification.

Observable dynamics, such as waves propagating on a surface, are generally governed by partial differential equations (PDEs), which are determined by the physical properties of the propagation media. The spatial variations of these properties lead to spatially dependent PDEs. It is useful in many fields to recover the variations from the observations of dynamical behaviors on the material. A method is proposed to form a map of the physical properties' spatial variations for a material via data-driven spatially dependent PDE identification and applied to recover acoustical properties (viscosity, attenuation, and phase speeds) for propagating waves. The proposed data-driven PDE identification scheme is based on ℓ1-norm minimization. It does not require any PDE term that is assumed active from the prior knowledge and is the first approach that is capable of identifying spatially dependent PDEs from measurements of phenomena. In addition, the method is efficient as a result of its non-iterative nature and can be robust against noise if used with an integration transformation technique. It is demonstrated in multiple experimental settings, including real laser measurements of a vibrating aluminum plate. Codes and data are available online at https://tinyurl.com/4wza8vxs.

Fastener partial looseness identification of the ballastless track by a time-domain Bayesian approach

The fastener system is an important component of the ballastless track, and the fastener looseness threatens the safety of the railway track. This paper proposes a Bayesian model updating method integrated with a Python–Abaqus interface framework to identify the fastener partial looseness of the ballastless track via vibration data. By following the technical standards of Chinese Railway Industry, a laboratory scaled ballastless track model was constructed for the demonstration and verification of the proposed methodology. Not only the damage location can be accurately identified, but also the damage severity can be evaluated. In addition, the associated posterior uncertainties of the identified results can be quantified by calculating the posterior probability density functions of all uncertain model parameters, which can provide valuable information to engineers for ballastless track system damage detection.

A nonparametric instrumental approach to confounding in competing risks models.

This paper discusses nonparametric identification and estimation of the causal effect of a treatment in the presence of confounding, competing risks and random right-censoring. Our identification strategy is based on an instrumental variable. We show that the competing risks model generates a nonparametric quantile instrumental regression problem. Quantile treatment effects on the subdistribution function can be recovered from the regression function. A distinguishing feature of the model is that censoring and competing risks prevent identification at some quantiles. We characterize the set of quantiles for which exact identification is possible and give partial identification results for other quantiles. We outline an estimation procedure and discuss its properties. The finite sample performance of the estimator is evaluated through simulations. We apply the proposed method to the Health Insurance Plan of Greater New York experiment.

Partial Identification in Matching Models for the Marriage Market

We study partial identification of the preference parameters in models of one-to-one matching with perfectly transferable utilities, without imposing parametric distributional restrictions on the unobserved heterogeneity and with data on one large market. We provide a tractable characterisation of the identified set, under various classes of nonparametric distributional assumptions on the unobserved heterogeneity. Using our methodology, we re-examine some of the relevant questions in the empirical literature on the marriage market which have been previously studied under the Multinomial Logit assumption.

Investigation of Partial Discharges within Power Oil Transformers by Acoustic Emission

This paper presents the authors’ multi-channel measurement systems designed and built to conduct research on partial discharge phenomena using the acoustic emission method. The systems provide real-time monitoring, recording of signals and analysis of recorded signals. The analysis is carried out in time, frequency, time-frequency and discrimination threshold domains. In particular, a descriptor with the ADC acronym is defined, which ranks the signals according to the so-called degree of advancement. Studies have been carried out, showing that for a single partial discharge source, when tested in parallel, using the electrical and acoustic emission methods, the ranking of the signals using this descriptor is identical to the ranking according to the value of the apparent charge introduced by sources. This paper presents the authors’ patented method of partial discharge location and identification in power oil transformers. The results of tests of power oil transformer at the test station, conducted in parallel with the electric method and the authors’ method, and the results of tests in three selected transformers carried out during ongoing in-situ operation using the authors’ method are presented. Based on these results, the authors make diagnoses of the condition of the insulation systems in the tested transformers. The inspections of these transformers confirm the diagnoses.