skip to main content

Title: Using the Standardized Root Mean Squared Residual (SRMR) to Assess Exact Fit in Structural Equation Models
We examine the accuracy of p values obtained using the asymptotic mean and variance (MV) correction to the distribution of the sample standardized root mean squared residual (SRMR) proposed by Maydeu-Olivares to assess the exact fit of SEM models. In a simulation study, we found that under normality, the MV-corrected SRMR statistic provides reasonably accurate Type I errors even in small samples and for large models, clearly outperforming the current standard, that is, the likelihood ratio (LR) test. When data shows excess kurtosis, MV-corrected SRMR p values are only accurate in small models ( p = 10), or in medium-sized models ( p = 30) if no skewness is present and sample sizes are at least 500. Overall, when data are not normal, the MV-corrected LR test seems to outperform the MV-corrected SRMR. We elaborate on these findings by showing that the asymptotic approximation to the mean of the SRMR sampling distribution is quite accurate, while the asymptotic approximation to the standard deviation is not.
; ;
Award ID(s):
Publication Date:
Journal Name:
Educational and Psychological Measurement
Page Range or eLocation-ID:
110 to 130
Sponsoring Org:
National Science Foundation
More Like this
  1. Summary This paper is concerned with empirical likelihood inference on the population mean when the dimension $p$ and the sample size $n$ satisfy $p/n\rightarrow c\in [1,\infty)$. As shown in Tsao (2004), the empirical likelihood method fails with high probability when $p/n>1/2$ because the convex hull of the $n$ observations in $\mathbb{R}^p$ becomes too small to cover the true mean value. Moreover, when $p> n$, the sample covariance matrix becomes singular, and this results in the breakdown of the first sandwich approximation for the log empirical likelihood ratio. To deal with these two challenges, we propose a new strategy of adding two artificial data points to the observed data. We establish the asymptotic normality of the proposed empirical likelihood ratio test. The proposed test statistic does not involve the inverse of the sample covariance matrix. Furthermore, its form is explicit, so the test can easily be carried out with low computational cost. Our numerical comparison shows that the proposed test outperforms some existing tests for high-dimensional mean vectors in terms of power. We also illustrate the proposed procedure with an empirical analysis of stock data.
  2. This study introduces the statistical theory of using the Standardized Root Mean Squared Error (SRMR) to test close fit in ordinal factor analysis. We also compare the accuracy of confidence intervals (CIs) and tests of close fit based on the Standardized Root Mean Squared Error (SRMR) with those obtained based on the Root Mean Squared Error of Approximation (RMSEA). We use Unweighted Least Squares (ULS) estimation with a mean and variance corrected test statistic. The current (biased) implementation for the RMSEA never rejects that a model fits closely when data are binary and almost invariably rejects the model in large samples if data consist of five categories. The unbiased RMSEA produces better rejection rates, but it is only accurate enough when the number of variables is small (e.g., p = 10) and the degree of misfit is small. In contrast, across all simulated conditions, the tests of close fit based on the SRMR yield acceptable type I error rates. SRMR tests of close fit are also more powerful than those using the unbiased RMSEA.
  3. Abstract Obtaining lightweight and accurate approximations of discretized objective functional Hessians in inverse problems governed by partial differential equations (PDEs) is essential to make both deterministic and Bayesian statistical large-scale inverse problems computationally tractable. The cubic computational complexity of dense linear algebraic tasks, such as Cholesky factorization, that provide a means to sample Gaussian distributions and determine solutions of Newton linear systems is a computational bottleneck at large-scale. These tasks can be reduced to log-linear complexity by utilizing hierarchical off-diagonal low-rank (HODLR) matrix approximations. In this work, we show that a class of Hessians that arise from inverse problems governed by PDEs are well approximated by the HODLR matrix format. In particular, we study inverse problems governed by PDEs that model the instantaneous viscous flow of ice sheets. In these problems, we seek a spatially distributed basal sliding parameter field such that the flow predicted by the ice sheet model is consistent with ice sheet surface velocity observations. We demonstrate the use of HODLR Hessian approximation to efficiently sample the Laplace approximation of the posterior distribution with covariance further approximated by HODLR matrix compression. Computational studies are performed which illustrate ice sheet problem regimes for which the Gauss–Newton data-misfit Hessianmore »is more efficiently approximated by the HODLR matrix format than the low-rank (LR) format. We then demonstrate that HODLR approximations can be favorable, when compared to global LR approximations, for large-scale problems by studying the data-misfit Hessian associated with inverse problems governed by the first-order Stokes flow model on the Humboldt glacier and Greenland ice sheet.« less
  4. Summary

    We introduce an L2-type test for testing mutual independence and banded dependence structure for high dimensional data. The test is constructed on the basis of the pairwise distance covariance and it accounts for the non-linear and non-monotone dependences among the data, which cannot be fully captured by the existing tests based on either Pearson correlation or rank correlation. Our test can be conveniently implemented in practice as the limiting null distribution of the test statistic is shown to be standard normal. It exhibits excellent finite sample performance in our simulation studies even when the sample size is small albeit the dimension is high and is shown to identify non-linear dependence in empirical data analysis successfully. On the theory side, asymptotic normality of our test statistic is shown under quite mild moment assumptions and with little restriction on the growth rate of the dimension as a function of sample size. As a demonstration of good power properties for our distance-covariance-based test, we further show that an infeasible version of our test statistic has the rate optimality in the class of Gaussian distributions with equal correlation.

  5. Context. Stars evolving along the asymptotic giant branch (AGB) can become carbon rich in the final part of their evolution. The detailed description of their spectra has led to the definition of several spectral types: N, SC, J, and R. To date, differences among them have been partially established only on the basis of their chemical properties. Aims. An accurate determination of the luminosity function (LF) and kinematics together with their chemical properties is extremely important for testing the reliability of theoretical models and establishing on a solid basis the stellar population membership of the different carbon star types. Methods. Using Gaia Data Release 2 ( Gaia DR2) astrometry, we determine the LF and kinematic properties of a sample of 210 carbon stars with different spectral types in the solar neighbourhood with measured parallaxes better than 20%. Their spatial distribution and velocity components are also derived. Furthermore, the use of the infrared Wesenheit function allows us to identify the different spectral types in a Gaia -2MASS diagram. Results. We find that the combined LF of N- and SC-type stars are consistent with a Gaussian distribution peaking at M bol  ∼ −5.2 mag. The resulting LF, however, shows two tails at lowermore »and higher luminosities more extended than those previously found, indicating that AGB carbon stars with solar metallicity may reach M bol  ∼ −6.0 mag. This contrasts with the narrower LF derived in Galactic carbon Miras from previous studies. We find that J-type stars are about half a magnitude fainter on average than N- and SC-type stars, while R-hot stars are half a magnitude brighter than previously found, although fainter in any case by several magnitudes than other carbon types. Part of these differences are due to systematically lower parallaxes measured by Gaia DR2 with respect to H IPPARCOS values, in particular for sources with parallax ϖ < 1 mas. The Galactic spatial distribution and velocity components of the N-, SC-, and J-type stars are very similar, while about 30% of the R-hot stars in the sample are located at distances greater than ∼500 pc from the Galactic plane, and show a significant drift with respect to the local standard of rest. Conclusions. The LF derived for N- and SC-type in the solar neighbourhood fully agrees with the expected luminosity of stars of 1.5−3 M ⊙ on the AGB. On a theoretical basis, the existence of an extended low-luminosity tail would require a contribution of extrinsic low-mass carbon stars, while the high-luminosity tail would imply that stars with mass values up to ∼5 M ⊙ may become carbon stars on the AGB. J-type stars differ significantly not only in their chemical composition with respect to the N- and SC-types, but also in their LF, which reinforces the idea that these carbon stars belong to a different type whose origin is still unknown. The derived luminosities of R-hot stars means that it is unlikely that these stars are in the red-clump, as previously claimed. On the other hand, the derived spatial distribution and kinematic properties, together with their metallicity values, indicate that most of the N-, SC-, and J-type stars belong to the thin disc population, while a significant fraction of R-hot stars show characteristics compatible with the thick disc.« less