Bias in the Estimation of Autocorrelations

Marriott, F. H. C.; Pope, J. A.

doi:10.2307/2332719

Cited by 149 publications

(181 citation statements)

References 0 publications

Supporting

Mentioning

175

Contrasting

Order By: Relevance

“…Third, and based on the results from the simulation study, if there are approximately five time points or less per phase, it is necessary to choose a more stringent nominal alpha to arrive at reliable results. Alternative approaches are applying a correction (Crosbie, 1993;Marriott & Pope, 1954) or resampling techniques (Edgington, 1992). One approach to resampling has been described by Borckardt et al (2008).…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating Statistical and Clinical Significance of Intervention Effects in Single-Case Experimental Designs: An SPSS Method to Analyze Univariate Data

et al. 2015

View full text Add to dashboard Cite

Single-case experimental designs are useful methods in clinical research practice to investigate individual client progress. Their proliferation might have been hampered by methodological challenges such as the difficulty applying existing statistical procedures. In this article, we describe a data-analytic method to analyze univariate (i.e., one symptom) single-case data using the common package SPSS. This method can help the clinical researcher to investigate whether an intervention works as compared with a baseline period or another intervention type, and to determine whether symptom improvement is clinically significant. First, we describe the statistical method in a conceptual way and show how it can be implemented in SPSS. Simulation studies were performed to determine the number of observation points required per intervention phase. Second, to illustrate this method and its implications, we present a case study of an adolescent with anxiety disorders treated with cognitive-behavioral therapy techniques in an outpatient psychotherapy clinic, whose symptoms were regularly assessed before each session. We provide a description of the data analyses and results of this case study. Finally, we discuss the advantages and shortcomings of the proposed method.

show abstract

Section: Discussionmentioning

confidence: 99%

“…This parameter may be underestimated for short time series (Marriott & Pope, 1954) and therefore tests on the parameters in model eq. 1 may remain too liberal (Crosbie, 1993).…”

Section: Accepted Manuscriptmentioning

confidence: 99%

Evaluating Statistical and Clinical Significance of Intervention Effects in Single-Case Experimental Designs: An SPSS Method to Analyze Univariate Data

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Remember that AR (and CR) effect estimates are subject to finite sample bias in general, hence, also in time-invariant models (Cheang & Reinsel, 2000;Marriott & Pope, 1954). The lower part of Table 5 therefore shows biases in a time-invariant VAR model fitted to a time series of length T = 50.…”

Section: Temporal Patterns In the Moderatormentioning

confidence: 96%

“…However, fixing the intercepts only (i.e., reducing the degrees of freedom in the estimation of the means) reduces bias in the AR and CR effects, but does not remove it. In fact, it has been shown that estimates of AR (and possibly CR) effects are biased toward 0 even with a known mean (Cheang & Reinsel, 2000;Marriott & Pope, 1954).…”

Section: Performance In Small Samplesmentioning

confidence: 99%

Capturing Context-Related Change in Emotional Dynamics via Fixed Moderated Time Series Analysis

Adolf

Voelkle

Brose

et al. 2017

Multivariate Behavioral Research

View full text Add to dashboard Cite

Much of recent affect research relies on intensive longitudinal studies to assess daily emotional experiences. The resulting data are analyzed with dynamic models to capture regulatory processes involved in emotional functioning. Daily contexts, however, are commonly ignored. This may not only result in biased parameter estimates and wrong conclusions, but also ignores the opportunity to investigate contextual effects on emotional dynamics. With fixed moderated time series analysis, we present an approach that resolves this problem by estimating context-dependent change in dynamic parameters in single-subject time series models. The approach examines parameter changes of known shape and thus addresses the problem of observed intra-individual heterogeneity (e.g., changes in emotional dynamics due to observed changes in daily stress). In comparison to existing approaches to unobserved heterogeneity, model estimation is facilitated and different forms of change can readily be accommodated. We demonstrate the approach's viability given relatively short time series by means of a simulation study. In addition, we present an empirical application, targeting the joint dynamics of affect and stress and how these co-vary with daily events. We discuss potentials and limitations of the approach and close with an outlook on the broader implications for understanding emotional adaption and development.

show abstract

“…The statistics in the table are computed from 10,000 samples generated from the AR(1) process r t+1 = θ 0 r t + u t+1 with T = 500, θ 0 = 0.4, 0.7, 0.95, and There are two main reasons for this simplification. First, it is well documented that this model could cause a large estimation bias as the persistence of the process increases and the bias properties of the OLS estimator are analytically derived and widely discussed in the literature (Marriott and Pope, 1954;Kendall, 1954; among others). Second, the i.i.d.…”

Section: Tablementioning

confidence: 99%