Hostname: page-component-7b7567c8c-wxvwp Total loading time: 0 Render date: 2026-11-18T06:22:01.752Z Has data issue: false hasContentIssue false

Estimation of Linear Models from Coarsened Observations: A Method of Moments Approach

Published online by Cambridge University Press:  10 March 2025

Bernard M. S. van Praag*
Affiliation:
Tinbergen Institute, University of Amsterdam, The Netherlands
J. Peter Hop
Affiliation:
Independent, The Netherlands
William H. Greene
Affiliation:
University of South Florida, USA
*
Corresponding author: Bernard M. S. van Praag; Email: b.m.s.vanpraag@uva.nl
Rights & Permissions [Opens in a new window]

Abstract

In the last few decades, the study of ordinal data in which the variable of interest is not exactly observed but only known to be in a specific ordinal category has become important. To emphasize that the problem is not specific to a specific discipline we will use the neutral term coarsened observation. For single-equation models estimation of the latent linear model by Maximum Likelihood (ML) is routine. But, for higher-dimensional multivariate models it is computationally cumbersome as estimation requires the evaluation of multivariate normal distribution functions on a large scale. Our proposed alternative estimation method, based on the Generalized Method of Moments (GMM), circumvents this multivariate integration problem. It can be implemented by repeated application of standard techniques and provides a simpler and faster approach than the usual ML approach. It is applicable to multiple-equation models with $K$-dimensional error correlation matrices and ${J}_k$ response categories for the kth equation. It also yields a simple method to estimate polyserial and polychoric correlations. Comparison of our method with the outcomes of the Stata ML procedure cmp yields estimates that are not statistically different, while estimation by our method requires only a fraction of the computing time.

Information

Type
Theory and Methods
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of Psychometric Society
Figure 0

Table 1In-between and full error covariance matrices for FMOP.

Figure 1

Table 2Regression estimates from FMOP and Ordered Probit.

Figure 2

Fig. 1A block of satisfaction questions with respect to various life domains.

Figure 3

Table 3Comparison of the parameter estimates and their s.e.’s for Ordered Probit, Method of Moments, and Maximum Likelihood.

Figure 4

Table 4Full error correlation matrices compared for SUOP and ML.

Figure 5

Table 5The polychoric correlation matrix.

Figure 6

Table 6The Pearson correlation matrix. (scale 0–10).

Figure 7

Table 7Beta’s, Standard errors and Error correlations (N = 10,000).

Figure 8

Table 8aBeta’s, Standard errors and Error correlations (N = 5,000) (Base dataset of 10000 cases, only every second case is used).

Figure 9

Table 8bBeta’s, Standard errors and Error correlations (N = 2,000) (Base dataset of 10000 cases, only every fifth case is used).

Figure 10

Table 8cBeta’s, Standard errors and Error correlations (N = 1,000) (Base dataset of 10000 cases, only every tenth case is used).

Figure 11

Table 8dBeta’s, Standard errors and Error correlations (N = 1,000) (A newly created dataset of 1000 cases).

Figure 12

Table 8eBeta’s, Standard errors and Error correlations (N = 1,000) (Again a newly created dataset of 1000 cases).

Figure 13

Table 8fThe correlation matrix of the variables (N = 10,000).

Figure 14

Table A1Relation between covariance $\rho$ and in-between covariance $\overline{\rho}$.

Figure 15

Figure. A1$\overline{\rho}$ as a function of $\rho$.

Figure 16

Table A2In-between and full covariance matrices for the five-year panel.

Figure 17

Table B1Comparison of the parameter estimates and their s.e.’s for Ordered Probit, Method of Moments, and Maximum Likelihood.