Questions tagged [diagnostic]

Diagnostic measures (such as residuals or some summary statistics calculated from residuals) are used to evaluate some aspect of quality of model fit to data.

Filter by
Sorted by
Tagged with
1
vote
0answers
46 views

Continuous fitted values for a categorical predictors in a mixed effects model and subsequent boot strapping

I have undertaken a reciprocal turf transplant in a blocked experimental design, where turfs have been moved in an out of high use areas with corresponding controls. I have measured the change in ...
0
votes
0answers
222 views

Partial residual plot not informative but transformations of predictor variables are required?

I have fitted a binary logistic regression model with two explanatory variables: $x_1,x_2$ using fractional polynomials (60k observations)(https://cran.r-project.org/web/packages/mfp/vignettes/...
10
votes
7answers
1k views

Is sensitivity or specificity a function of prevalence?

Standard teaching says that sensitivity and specificity are properties of the test and are independent of prevalence. But isn't this just an assumption? Harrison's principles of internal medicine ...
3
votes
1answer
161 views

Linear Model Diagnostics via Machine Learning

Are there any approaches to checking diagnostics of statistical models, in particular linear regression, by machine learning methods? Many of the standard frequentist tests for numerical estimation of ...
0
votes
1answer
39 views

How should age be handled when obtaining a diagnostic biomarker signature?

I wonder if anyone can advise me on how the issue of age should be handled when identifying diagnostic biomarkers, combining them (e.g. via logistic regression, lasso) to obtain a signature to improve ...
4
votes
0answers
2k views

Influence plot for potential outlier detection from logistic regression in R

I am looking into identifying extreme values from their contribution to a binary outcome model. I have an unbalanced set and some extreme values which are part of the smaller set to predict (i.e ...
4
votes
1answer
1k views

Beta regression and regression diagnostics. Do we need to check for normality and other diagnostics?

I have a dependent variable which is a ratio and 0 < y < 1 condition holds. I will apply betareg in Stata but I am not sure what are the diagnostics that are ...
0
votes
1answer
186 views

What to do if a GARCH model selected by BIC has correlated residuals?

Suppose I fit a GARCH(1,1) model with Student-$t$ innovations of the standardized residuals using BIC selection. My mean model is ARMA(0,0). What can I do when the standardized residuals are still ...
1
vote
1answer
83 views

Comment on fitness of multi linear model

I have a dataset which I ran the lm() below. ...
1
vote
2answers
3k views

What is the usefulness of detection rate in a confusion matrix?

In the R caret documentation for confusionMatrix(): ...
2
votes
0answers
337 views

What do people mean when they say dfbetas in glms

I will follow the notation in the following article in this question Williams, D. (1987). Generalized Linear Model Diagnostics Using the Deviance and Single Case Deletions. Journal of the Royal ...
1
vote
0answers
33 views

Statistical pre-checks on transaction data (real estate)

I'm currently working on a paper in the field of real estate. As it's been some time since my last quant finance class, I just wanted to cross check my understanding of the problem at hand. The data ...
0
votes
1answer
3k views

Ljung-Box test for squared residuals values interpretation

I ran the Ljung-Box for a single series and find that the statistic is very high. I am using 20 lags so the critical value is 31.4104 and the statistic is greater than that. So my conclusion is that ...
1
vote
3answers
1k views

Application of Pareto/NBD and Pareto/GGG models for customer lifetime value estimates in high churn setting

I have been attempting to estimate customer lifetime value in the context of online classifieds (high churn context) using probabilistic models, chiefly the Pareto/NBD and Pareto/GGG techniques ...
1
vote
1answer
555 views

Diagnosing linearity for multiple regression model

I have the following model in R to estimate the change in mosquito abundance each year when controlling for climate variables: ...
1
vote
0answers
334 views

Multiple linear regression on time series data, is there autocorrelation of residuals?

I am applying a multiple linear regression model on time series data (independent and all dependent variables are time series). I am wondering if I have concerns of autocorrelation of residuals, by ...
5
votes
2answers
1k views

Diagnostics for General Linear Models

Pearson residuals follow normal distribution. We plot them against predicted values to see if the model is good. Why would we plot deviance residuals against predicted values? Deviance residuals don'...
1
vote
1answer
117 views

meta-analysis for area under the ROC curve

I have a question on diagnostic test accuracy (DTA) meta-analysis. I understand that the current practice is to use either bivariate or HSROC method. Wondering, why can't we just pool the area ...
0
votes
0answers
604 views

Bootstrapping to reduce influence of outliers

I've fit a regression model and diagnostic tests are showing some values have high Cook's D as well as high DFBETAS on my parameter of interest. The effect does not cross the traditional threshold for ...
2
votes
1answer
112 views

Distribution of residuals envelops normal distribution

when plotting the distribution of the obtained residuals (from a regression) in stata to see if they fit a normal distribution, I get following result: This seemed very odd to me at first but when I ...
1
vote
2answers
2k views

What is a good number of treedepth saturations for a fit stan model?

I recently ran a model with this output: ...
1
vote
0answers
662 views

What to do with a downward band in the Residuals vs Fitted plot from linear regressions?

I got a residuals vs fitted plot with a clear downward pattern from a linear regression model and am wondering what are some of the sensible things to do in this situation? Doing a boxcox ...
0
votes
1answer
248 views

rstan: Diagnostics of regression

I ran a simple normal regression in rstan with some informative priors. My data has heteroskedasticity and would like to fix the same. However, am new to bayesian regression and rstan. My questions ...
2
votes
0answers
374 views

How do you get quantile normalized residuals for a t-distribution fit?

I've fitted a non-exponential family GLM regression model with the response distributed as a t-distribution with $\nu$ degrees of freedom, scale $\theta$ and mean $\mu = X\beta$. We estimate $\beta,\...
0
votes
1answer
651 views

What is the residual distribution in this regression model?

I have fitted a model $$g(\mu) = X\beta, \ \mu = EY_i, \ Y \sim T(\nu, \mu, \sigma)$$ where $T$ is a scaled and shifted t-distribution. This model was fitted in R (I use gam from mgcv), and pearson ...
0
votes
0answers
190 views

AIC versus model diagnostics, which one to go by?

I fitted model 1 and model 2. I then plotted the residual plots. Model 1 had some minor problems with the residual plot: there was more variance in the residuals on the right side than on the left. ...
3
votes
1answer
134 views

Why is there no more work done on convergence diagnosis of MCMC?

Most of the relevant papers and reviews date all the way back to 2003 or 1998. I can barely find anything newer. What makes it losing interest among researchers? I thought monitoring convergence for ...
2
votes
1answer
609 views

Diagnostic accuracy meta-analysis using MADA in R

After using the bivariate model I got a pooled sensitivity and false positive rate. How do I calculate the likelihood ratios, specificity etc with confidence intervals from here on? Can it be done ...
1
vote
1answer
139 views

Fitted values and diagnostics in MCMC model

I'm fitting some GLM models with different link functions (logit, probit and cloglog) with JAGS package. I have no experience at all with MCMC based models then I have three main doubts: 1) After I ...
4
votes
1answer
914 views

Serial correlation in financial returns (S&P500) in large samples

I have a daily stock returns series and a squared returns series for the S&P500. The Ljung-Box Q-Statistic for the squared returns series says there is autocorrelation and therefore ARCH effects ...
2
votes
1answer
1k views

Remaining heteroskedasticity even after GARCH estimation

This is according to the Ljung-Box $Q$ statistic of residuals squared and ARCH-LM test. Both suggest there are ARCH effects remaining after lag 1 even after I have estimated my GARCH (1,1) model. I ...
3
votes
1answer
616 views

Procedure of PRESS statistic

It is not very clear for me the calculation steps of PRESS statistic. What I have found: 1)we set aside the $j_{th}$ observation $⟨x_j,y_j⟩$ from the training set(It means, that we just remove the ...
2
votes
1answer
2k views

Poisson regression residuals diagnostic

I have done a poisson regression on my data set and am now looking to investigate the model fit. I notice that the fitted values from predict() in r give me the pre exp transformed values. For ...
0
votes
0answers
391 views

Model with piecewise-linear functions

I have a model which is of the following deterministic form: Y = A + B + C A, B and C are piecewise-linear functions of a, b and c, which are macroeconomic variables. I have the values (nodes, ...
1
vote
0answers
182 views

Influence function for glm?

I would like to calculate the influence of each sample on the coefficient under a logistic regression model. The R built-in function influence() is suppose to do ...
0
votes
1answer
30 views

Book on diagnostic research

Other than the excellent book Statistical Methods in Diagnostic Medicine, are there any other reference books on the argument? If not what are the best book which could be related to diagnostic ...
1
vote
1answer
1k views

Apparent correlation between standardized residuals and predicted in regression

The following is the residuals vs predicted scatter plot for a regression model with two IVs. Initially, I thought it was evidence of heteroskedasticity. But, I reasoned that although there is a ...
3
votes
0answers
308 views

Traceplots and gelman.rubin statistics for analysing convergence of mixtures of discrete and continuous distributions

I have a Bayesian hierarchical model which contains a number of distributions which are mixtures of a point-mass at zero and a continuous random variable. The model is fitted using a gibbs sampler. ...
0
votes
0answers
25 views

How to describe an error that occurs when your reference standard is falsely positive?

Here's sort of a TL;DR: how would you call it, if you incorrectly say your test was "false negative", because in fact the reference standard got it wrong (and was false positive). So my new test is ...
4
votes
0answers
76 views

Ill-behaved, nonnormal residuals of multiple regression: should I be concerned?

I have data on waist circumference (cm) (waist), gender, age and physical activity (vigorous MET minutes per week) (PA). I was trying to run linear regression in R on the model waist ~ gender + age + ...
0
votes
1answer
167 views

Diagnostic test

I want to calculate the sensitivity, specificity, positive predictive value and negative predictive value of two diagnostic tests. The sensitivity is 0.92 (95% CI, 0.82-1.03) Is this confidence ...
1
vote
1answer
425 views

Alternative tests for determining whiteness of a residual time series that does not result from an ARIMA process?

I’m doing some work that involves time series analysis of climate data, and I’m trying to figure out the best way to test for whiteness in a residual time series. In the course of this work I’m ...
2
votes
1answer
82 views

Contrast if regression coefficients of two predictors are equals

I've got this model: model <- lm(biomass ~ salin+ph+k+na+zn, data = sparData) with sparData dataset: ...
0
votes
1answer
621 views

Geweke Diagnostics in Bayesian regression model

I'm a newbie in Bayesian modelling and trying to understand something more on such field by running a Poisson regression and analyzing count data. Browsing on the internet, I found a set of ...
4
votes
1answer
3k views

Cox Snell residuals in R

I'm trying to fit parametric survival models for a data that I have, and I don't know how to get the Cox Snell residuals in R. An example of dataset with Exponential model fitted ...
1
vote
1answer
4k views

Bartlett test error (at least 2 observations) to check homoscedasticity

I've got this data.frame (framingham) sex sbp bmi 1 female 130 24.0 2 male 126 29.0 3 female 146 26.6 4 male 116 20.8 5 female 100 20.9 6 female 128 26.8 ...
1
vote
0answers
351 views

How to run regression diagnostics in big data setting

Many statistical software has powerful tools to help us to run regression diagnostics. For example, in R, we have many diagnostic plots for regression model (residuals vs fitted, QQ plot on residuals, ...
2
votes
1answer
223 views

Diagnostics in smoothing splines

I'm studying about Smoothing Splines and I'm having some doubts about this method. I already understood the criterion to choose the smooth parameter, but How I acess the fit of this type of non-...
1
vote
1answer
172 views

Should I analyze the residuals of a model selected by auto.arima?

I am a PhD student working on time series forecasting using neural networks and genetic algorithms. My question is: Is it necessary to analyze the residue, if I use the ...
3
votes
0answers
335 views

Observations aligned on PCA diagnostic plot: geometric interpretation?

The image below depicts the distance-distance plot for a (robust) PCA fit of a real data set. The distance-distance plot is described in greater detail on page 30-31 of (1) or page 2--3 of (0). It ...