Drawing causal inference with observational studies is the central pillar of many disciplines. One sufficient condition for identifying the causal effect is that the treatment-outcome relationship is unconfounded conditional on the observed covariates. It is often believed that the more covariates we condition on, the more plausible this unconfoundedness assumption is. This belief has had a huge impact on practical causal inference, suggesting that we should adjust for all pretreatment covariates. However, when there is unmeasured confounding between the treatment and outcome, estimators adjusting for some pretreatment covariate might have greater bias than estimators without adjusting for this covariate. This kind of covariate is called a bias amplifier, and includes instrumental variables that are independent of the confounder, and affect the outcome only through the treatment. Previously, theoretical results for this phenomenon have been established only for linear models. We fill in this gap in the literature by providing a general theory, showing that this phenomenon happens under a wide class of models satisfying certain monotonicity assumptions. We further show that when the treatment follows an additive or multiplicative model conditional on the instrumental variable and the confounder, these monotonicity assumptions can be interpreted as the signs of the arrows of the causal diagrams.

It is important to draw causal inference from observational studies, but this becomes challenging if the confounders have missing values. Generally, causal effects are not identifiable if the confounders are missing not at random. In this article we propose a novel framework for nonparametric identification of causal effects with confounders subject to an outcome-independent missingness, which means that the missing data mechanism is independent of the outcome, given the treatment and possibly missing confounders. We then propose a nonparametric two-stage least squares estimator and a parametric estimator for causal effects.

This Letter reports new results on muon neutrino disappearance from NOvA, using a 14 kton detector equivalent exposure of 6.05×10^{20} protons on target from the NuMI beam at the Fermi National Accelerator Laboratory. The measurement probes the muon-tau symmetry hypothesis that requires maximal θ_{23} mixing (θ_{23}=π/4). Assuming the normal mass hierarchy, we find Δm_{32}^{2}=(2.67±0.11)×10^{-3}  eV^{2} and sin^{2}θ_{23} at the two statistically degenerate values 0.404_{-0.022}^{+0.030} and 0.624_{-0.030}^{+0.022}, both at the 68% confidence level. Our data disfavor the maximal mixing scenario with 2.6σ significance.

Using two years of data from the NOvA Near Detector at Fermilab, we report a seasonal variation of cosmic ray induced multiple-muon (Nμ≥2) event rates which has an opposite phase to the seasonal variation in the atmospheric temperature. The strength of the seasonal multiple-muon variation is shown to increase as a function of the muon multiplicity. However, no significant dependence of the strength of the seasonal variation of the multiple-muon variation is seen as a function of the muon zenith angle, or the spatial or angular separation between the correlated muons.

We report results from the first search for sterile neutrinos mixing with active neutrinos through a reduction in the rate of neutral-current interactions over a baseline of 810 km between the NOvA detectors. Analyzing a 14-kton detector equivalent exposure of 6.05×1020 protons-on-target in the NuMI beam at Fermilab, we observe 95 neutral-current candidates at the Far Detector compared with 83.5±9.7(stat)±9.4(syst) events predicted assuming mixing only occurs between active neutrino species. No evidence for νμ→νs transitions is found. Interpreting these results within a 3+1 model, we place constraints on the mixing angles θ24<20.8° and θ34<31.2° at the 90% C.L. for 0.05 eV2≤Δm412≤0.5 eV2, the range of mass splittings that produce no significant oscillations over the Near Detector baseline.

The NOvA experiment has seen a 4.4σ signal of ν[over ¯]_{e} appearance in a 2 GeV ν[over ¯]_{μ} beam at a distance of 810 km. Using 12.33×10^{20} protons on target delivered to the Fermilab NuMI neutrino beamline, the experiment recorded 27 ν[over ¯]_{μ}→ν[over ¯]_{e} candidates with a background of 10.3 and 102 ν[over ¯]_{μ}→ν[over ¯]_{μ} candidates. This new antineutrino data are combined with neutrino data to measure the parameters |Δm_{32}^{2}|=2.48_{-0.06}^{+0.11}×10^{-3}  eV^{2}/c^{4} and sin^{2}θ_{23} in the ranges from (0.53-0.60) and (0.45-0.48) in the normal neutrino mass hierarchy. The data exclude most values near δ_{CP}=π/2 for the inverted mass hierarchy by more than 3σ and favor the normal neutrino mass hierarchy by 1.9σ and θ_{23} values in the upper octant by 1.6σ.

The two-detector design of the NOvA neutrino oscillation experiment, in which two functionally identical detectors are exposed to an intense neutrino beam, aids in canceling leading order effects of cross-section uncertainties. However, limited knowledge of neutrino interaction cross sections still gives rise to some of the largest systematic uncertainties in current oscillation measurements. We show contemporary models of neutrino interactions to be discrepant with data from NOvA, consistent with discrepancies seen in other experiments. Adjustments to neutrino interaction models in GENIE are presented, creating an effective model that improves agreement with our data. We also describe systematic uncertainties on these models, including uncertainties on multi-nucleon interactions from a newly developed procedure using NOvA near detector data.

Results are reported from an improved measurement of ν_{μ}→ν_{e} transitions by the NOvA experiment. Using an exposure equivalent to 6.05×10^{20} protons on target, 33 ν_{e} candidates are observed with a background of 8.2±0.8 (syst.). Combined with the latest NOvA ν_{μ} disappearance data and external constraints from reactor experiments on sin^{2}2θ_{13}, the hypothesis of inverted mass hierarchy with θ_{23} in the lower octant is disfavored at greater than 93% C.L. for all values of δ_{CP}.

The cross section of neutrino-induced neutral-current coherent π0 production on a carbon-dominated target is measured in the NOvA near detector. This measurement uses a narrow-band neutrino beam with an average neutrino energy of 2.7 GeV, which is of interest to ongoing and future long-baseline neutrino oscillation experiments. The measured, flux-averaged cross section is σ=13.8±0.9(stat)±2.3(syst)×10-40 cm2/nucleus, consistent with model prediction. This result is the most precise measurement of neutral-current coherent π0 production in the few-GeV neutrino energy region.

Using the NOvA neutrino detectors, a broad search has been performed for any signal coincident with 28 gravitational wave events detected by the LIGO/Virgo Collaboration between September 2015 and July 2019. For all of these events, NOvA is sensitive to possible arrival of neutrinos and cosmic rays of GeV and higher energies. For five (seven) events in the NOvA Far (Near) Detector, timely public alerts from the LIGO/Virgo Collaboration allowed recording of MeV-scale events. No signal candidates were found.