In many applications of functional data analysis, summarising functional variation based on fits, without taking account of the estimation process, runs the risk of attributing the estimation variation to the functional variation, thereby overstating the latter. For example, the first eigenvalue of a sample covariance matrix computed from estimated functions may be biased upwards. We display a set of estimated neuronal Poisson-process intensity functions where this bias is substantial, and we discuss two methods for account ing for estimation variation. One method uses a random-coefficient model, which requires all functions to be fitted with the same basis functions. An alternative method removes the same-basis restriction by means of a hierarchical Gaussian process model. In a small simulation study the hierarchical Gaussian process model outperformed the random coefficient model and greatly reduced the bias in the estimated first eigenvalue that would result from ignoring estimation variability. For the neuronal data the hierarchical Gaussian process estimate of the first eigenvalue was much smaller than the naive estimate that ignored variability due to function estimation. The neuronal setting also illustrates the benefit of incorporating alignment parameters into the hierarchical scheme.
This article presents two methods of testing the hypothesis of equality of two functions H(0):f(1)(t)=f(2)(t) for all t, in a generalized non-parametric regression framework using a recently developed generalized non-parametric regression method called Bayesian adaptive regression splines (BARS). Of particular interest is the special case of testing equality of two Poisson process intensity functions lambda(1) (t)=lambda(2) (t), which arises frequently in neurophysiological applications. The first method uses Bayes factors, and the second method uses a modified Hotelling T(2) test. Both methods are applied to the analysis of 347 motor cortical neurons and, for certain choices of test criteria, the two methods lead to the same conclusions for all but 7 neurons. A small simulation study of power indicates that the Bayes factor can be somewhat more powerful in small samples. The T(2)-type test should be useful in screening large number of neurons for condition-related activity, while the Bayes factor will be especially helpful in assessing evidence in favour of H(0).
When correlation is measured in the presence of noise, its value is decreased. In single-neuron recording experiments, for example, the correlation of selectivity indices in a pair of tasks may be assessed across neurons, but, because the number of trials is limited, the measured index values for each neuron will be noisy. This attenuates the correlation. A correction for such attenuation was proposed by Spearman more than 100 yr ago, and more recent work has shown how confidence intervals may be constructed to supplement the correction. In this paper, we propose an alternative Bayesian correction. A simulation study shows that this approach can be far superior to Spearman's, both in accuracy of the correction and in coverage of the resulting confidence intervals. We demonstrate the usefulness of this technology by applying it to a set of data obtained from the frontal cortex of a macaque monkey while performing serial order and variable reward saccade tasks. There the correction results in a substantial increase in the correlation across neurons in the two tasks.
Coccidioidomycosis (Valley Fever) is a fungal infection found in the southwestern United States, northern Mexico, and some places in Central and South America. The fungi that cause it (Coccidioides immitis and Coccidioides posadasii) are normally soil dwelling, but, if disturbed, become airborne and infect the host when their spores are inhaled. It is thus natural to surmise that weather conditions, which foster the growth and dispersal of Coccidioides, must have an effect on the number of cases in the endemic areas. This article reviews our attempts to date at quantifying this relationship in Kern County, California (where C. immitis is endemic). We have examined the effect on incidence resulting from precipitation, surface temperature, and wind speed. We have performed our studies by means of a simple linear correlation analysis, and by a generalized autoregressive moving average model. Our first analysis suggests that linear correlations between climatic parameters and incidence are weak; our second analysis indicates that incidence can be predicted largely by considering only the previous history of incidence in the county-the inclusion of climate-or weather-related time sequences improves the model only to a relatively minor extent. Our work therefore suggests that incidence fluctuations (about a seasonally varying background value) are related to biological and/or anthropogenic reasons, and not so much to weather or climate anomalies.
Coccidioidomycosis (valley fever) is a fungal infection found in the southwestern US, northern Mexico, and some places in Central and South America. The fungus that causes it (Coccidioides immitis) is normally soildwelling but, if disturbed, becomes air-borne and infects the host when its spores are inhaled. It is thus natural to surmise that weather conditions that foster the growth and dispersal of the fungus must have an effect on the number of cases in the endemic areas. We present here an attempt at the modeling of valley fever incidence in Kern County, California, by the implementation of a generalized auto regressive moving average (GARMA) model. We show that the number of valley fever cases can be predicted mainly by considering only the previous history of incidence rates in the county. The inclusion of weather-related time sequences improves the model only to a relatively minor extent. This suggests that fluctuations of incidence rates (about a seasonally varying background value) are related to biological and/or anthropogenic reasons, and not so much to weather anomalies.
We propose a scalable semiparametric Bayesian model to capture dependencies among multiple neurons by detecting their co-firing (possibly with some lag time) patterns over time. After discretizing time so there is at most one spike at each interval, the resulting sequence of 1’s (spike) and 0’s (silence) for each neuron is modeled using the logistic function of a continuous latent variable with a Gaussian process prior. For multiple neurons, the corresponding marginal distributions are coupled to their joint probability distribution using a parametric copula model. The advantages of our approach are as follows: the nonparametric component (i.e., the Gaussian process model) provides a flexible framework for modeling the underlying firing rates; the parametric component (i.e., the copula model) allows us to make inference regarding both contemporaneous and lagged relationships among neurons; using the copula model, we construct multivariate probabilistic models by separating the modeling of univariate marginal distributions from the modeling of dependence structure among variables; our method is easy to implement using a computationally efficient sampling algorithm that can be easily extended to high dimensional problems. Using simulated data, we show that our approach could correctly capture temporal dependencies in firing rates and identify synchronous neurons. We also apply our model to spike train data obtained from prefrontal cortical areas.
We consider the problem of comparing noisy functions, here trial-averaged neuronal firing-rate curves, across multiple experimental conditions. Of interest are comparisons both within neurons and also among populations of individually recorded neurons. We propose likelihood ratio tests to perform comparisons either pointwise or globally over the entire experimental time. A simulation study of power demonstrates the strength of these tests even for moderate sample sizes. We implement these tests on a group of 233 neurons recorded from primate frontal oculomotor cortex, first, to screen for condition-related differential activity and, second, to search for neurons displaying interesting time-locked features that vary with condition.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.