Questions tagged [sensitivity-analysis]
Auxiliary methods intended to check if the outcome of an analysis strongly depends on the model assumptions, preprocessing steps, presence of outliers, etc.
164 questions
2
votes
0
answers
19
views
Sensitivity Analysis with percentage-type output
I'm trying to calibrate a model, but I wanted to take a step back and perform a sensitivity analysis to assess if the parameters I'm calibrating are actually influencing the output.
The simplest ...
2
votes
0
answers
46
views
Can Bayesian Priors Be Used to Address Unadjusted Confounding?
Imagine I conduct a selection on observables identification strategy using regression adjustment and regress the outcome on the exposure of interest plus ten identified confounders.
Then, suppose I ...
- 161
0
votes
0
answers
17
views
Why functional GSA and not piece wise?
One can ran a Global Sensitivity Analysis in R using packages like {sensitivity}, {sensobol} or {BASS}. sensitivity::sensiHSIC() and BASS can do GSA on functional responses (and scalar predictors).
I'...
- 528
0
votes
0
answers
129
views
Post-hoc sensitivity analysis for multiple logistic regression?
A reviewer is asking about whether we conducted an a priori power analysis (we did not - mistake, I know). The senior author wants us to address this reviewer's question by performing a post-hoc ...
- 71
1
vote
0
answers
45
views
Why does the Sobol Sensitivity (S1) so large in my model [closed]
I am running a model in Julia with 42 parameters (they are somehow inner-correlated with each other) and 6 outputs, and trying to use the function gsa from the ...
- 213
0
votes
0
answers
96
views
Aggregating (partial rank) correlation coefficients
I am doing sensitivity analysis of an epidemic model per
Blower 1994, namely:
Latin hypercube sampling
(LHS)
from parameter (model input) priors, then
Partial Rank Correlation Coefficients
(PRCC)
for ...
- 301
2
votes
1
answer
88
views
In NLLS, how do you produce accurate estimates of RMSE(true_params) given RMSE(global_minimum_params)?
I have an exponential decay
$f(t) = \sum_n \left( A_n e^{-\frac{t}{\tau_n}} \right) + c + \epsilon(t)$,
where n represents the different exponential decay components, $A_n$ represents each decay ...
- 21
1
vote
0
answers
98
views
Interpreting the plot for sensitivity analysis for mediation analysis using the mediation package
I have performed a mediation analysis using the mediation package for R and the recommended sensitivity analysis. Despite reading "Identification, Inference, and Sensitivity Analysis for Causal ...
0
votes
1
answer
127
views
Bayes Factor for two exact hypotheses on normally distributed data with unknown variance
Let's assume that the observations $x_i$ are normally distributed
$$
x_i \sim N(\mu, \sigma^2)
$$
and that the variance $\sigma^2$ is unknown. The Bayes Factor to compare two point hypotheses on the ...
- 63
1
vote
0
answers
36
views
Correlated parameters
If I use a regression model for estimating a time-to-event distribution function with several parameters (such as a Weibull), should I consider that those parameters are correlated when I later ...
- 43
1
vote
1
answer
92
views
Rank the sensitivity among multiple calibration curves
Let's say that, with a measurement device, we have a linear relationship between an output measurement I (in mV) and the concentration ...
- 632
1
vote
0
answers
132
views
Sensitivity analysis of features in a Random Forest Model
I have a built a large Random Forest Classifier and was able to output the feature importance as below:
I understand that this importance is a based on mean decreased impurity. But how to interpret ...
- 11
0
votes
0
answers
87
views
Global sensitivity indices with sign (positive/negative)
I’ve used Sobol global sensitivity indices to analyze the effect that some input variables have on a nonlinear regression model, but now I need to know if the effect is positive or negative (i.e. if ...
- 497
1
vote
0
answers
134
views
Sensitivity Analysis for Raw Model Output or Model Predictions?
For a variety of reasons, many researchers have suggested that, for attempting to make causal inferences with non-linear statistical models, one should generally avoid endowing a causal interpretation ...
- 161
3
votes
1
answer
86
views
Nonlinearity of model using Sobol indices
I'm analyzing a computationally demanding numeric model where I want to show that nonlinearities play a certain role for my problem. I want to do this using Sobol sensitivity indices of first oder by ...
- 61