Questions tagged [regression-strategies]

Regression Modeling Strategies

Filter by
Sorted by
Tagged with
0
votes
1answer
250 views

Variable with predictor name in Predict.rms

I would like to build regression models for a number of different exposure variables and show the results using ggplot.Predict of the ...
1
vote
2answers
414 views

Wrong predictions with imbalanced binary data

I have a sample of 3000 observations. I want to study the impact of covariates on a binary dependent variable (i.e. two categories: "yes" or "no") using either a logistic regression or a linear ...
0
votes
1answer
36 views

linear model using significant outcomes of linear model

I am getting confused about a couple of techniques that I'd like to discuss: I am using an R package that in turn depends on the limma package. I tried reading the docs, but I can't grasp what is ...
0
votes
1answer
159 views

How to set up a regression for Adjusted Plus Minus with no offense and defense?

So here is the backstory: This is a game with two teams The outcome of the game is that one of the teams wins and the other loses There is no offense and defense We can use tug of war as an analogy (...
0
votes
1answer
70 views

How to decide the size of the subset of the most relevant features while performing feature selection?

I have a dataset with 25 features(columns). I'm trying to apply forward feature selection to my dataset, which will return a subset containing the best features. But the size of this subset, i.e. the ...
1
vote
0answers
57 views

Including many interaction terms in a logistic regression

I have a logistic regression model that is currently of the form: Event ~ Vacc + Age I want to start including interaction terms for different types of vaccine. ...
1
vote
0answers
50 views

Interpretation of spline models: plot vs estimate

I have conducted regression analyses (cph - using rms package) to investigate associations between an exposure (air pollutant) ...
0
votes
1answer
453 views

machine-learning : Why training set and test set need to be independent and identically distributed?

My machine-learning book that I'm reading only says that they need to be but not why? My intuition says that if they are that leads to a better learning, if they were not it would be like we are ...
0
votes
1answer
85 views

Quantiles in ols/cph

I need to conduct regression analyses with continuous and categorised data (as the latter is common in my field). So far, I used to do this as follows: ...
3
votes
3answers
651 views

How can I find the correlation between groups of *attributes*?

Assume I have data where multiple attributes are measured for countries and the attributes can be divided into dimensions. For example one dimension can be 'Education' and have 5 attributes associated ...
2
votes
1answer
493 views

For LASSO, what is the best alternative when a held out test set is not viable?

I have a very limited data set with a number of features $\textbf{x}$ (roughly 30 dimensions) and a regression target $y$. I have roughly $20$ groups of data-points with $2$ points in each group (...
3
votes
4answers
910 views

What is the precise definition of unsupervised learning?

Let's look at a special case: Generative Adversarial Networks (GANs). (For those who don't know what a GAN is: for this purpose they are two neural networks that are trained using user generated ...
9
votes
1answer
2k views

How to estimate a calibration curve with bootstrap (R)

Question: I have fitted a probabilistic model (bayesian network) for modeling a binary outcome variable. I would like to create a high-resolution calibration plot (e.g. spline) corrected for ...
0
votes
1answer
318 views

Internal & External validation of ordinal outcome

For binary outcomes, I have used bootstrapping approach (Frank Harrell's R package rms) for internal validation and Cox framework for external validation. More ...
5
votes
1answer
288 views

Determine rate of change in dissimilarity (distance)?

I have repeated measures plant abundance data for 36 forest plots, across 80 years involving 50+ species of trees. The data are structured as: Columns = ...
2
votes
0answers
1k views

Confusing aspect of Harrell's bootstrap-based optimism-correction Internal Validation procedure

This bootstrap-based internal validation procedure is implemented in the validate function in Harrell's rms package. It allows ...
24
votes
2answers
2k views

Bayesian thinking about overfitting

I've devoted much time to development of methods and software for validating predictive models in the traditional frequentist statistical domain. In putting more Bayesian ideas into practice and ...
0
votes
1answer
1k views

How to plot the calibration curve for an ordinal logistic regression model applied to a test sample?

I'm doing a validation study of an ordinal logistic regression model that was made with the lrm function of the rms package in R. How can I plot the calibration curve for the model when applied to new ...
0
votes
0answers
22 views

Standardised rates

I'm having trouble understanding the common practice of standardisation in ecological studies involving mortality/incident rates. Based on this paper by Jeffrey Milyo and Jennifer M Mellor advocates ...
0
votes
1answer
94 views

How to model environmental factors in a regression - and how does it not introduce multicolinearity in this specific case?

I am reading an analysis which takes a model that regresses a response variable (= criminal activity) on a socio-economic independent variable (= parent salary during childhood). The parameters are ...
0
votes
1answer
54 views

Is there a acceptable replacement for Poisson regression for this case?

I am very much a novice. I am trying to model a data that contains non-integers and zero values. The data contains the number of prescriptions written per county versus the number of overdose ...
1
vote
0answers
200 views

Why is the penalty in the logistic regression likelihood ratio test different from the penalty I specified when fitting the model?

I'm fitting a penalized logistic regression model using the rms package in R. When I print the result, the penalty in the model likelihood ratio test is different from the penalty I used to fit the ...
0
votes
1answer
262 views

Obtaining the odds ratio of a 2x2 logistic regression interaction in rms

I am using the RMS package and attempted to obtain the adjusted odds ratio for a logistic regression interaction. There does not seem to be a straightforward way to do this. I have decided to obtain ...
6
votes
1answer
2k views

Reporting the effect of a predictor in a logistic regression fitted with a restricted cubic spline

I have been playing around with using restricted cubic splines using the RMS package. Output below. ...
1
vote
0answers
703 views

package 'rms' - singular information matrix encountered

I am trying to fit a binary logistic regression model using only one predictor. After consulting a loess plot I decided to use restricted cubic spline to fit the curve (with default 4 knots). Using '...
0
votes
1answer
56 views

Seeking regression modeling strategies for predicting prices based on categorical variables (one of which is ordered)

I have a question similar to this one, which never received an answer. Let's say I have widgets that have different quality ratings $q\in\{0,1,\dots,N_q\}$ and which are in different regions $R\in\{...
2
votes
0answers
704 views

Are variable selection strategies for regression useful?

I have read many posts (including Frank Harrell's book) about the consequences of using variable selection strategies. However, it seems that many of the published work in the medical field still ...
1
vote
0answers
70 views

Can I treat ordinal values (with an underlying order) as continuous when extracting residuals and in exploratory factor analysis?

I am looking for clarification on whether it is acceptable to treat ordinal values with an underlying order as continuous when extracting residuals and in exploratory factor analysis (EFA). Does ...
3
votes
0answers
6k views

Family in GLM - how to choose the right one?

When modeling data sampled in the field, I often come across the problem of determining the Family of the dependent variable for GLM (or GLMM). An example: in an ecological study, I have ~ 60 patches. ...
2
votes
1answer
79 views

Setting contrasts for 10-level categorical variable

I have survey data on income and support for environmental protection. Income is a continuous variable that I have broken up into deciles. I have a hypothesis that support for protection ('Agree') ...
6
votes
1answer
2k views

Ordinal regression: logit, probit, complementary log-log or negative log-log?

Unfortunately, I found only this paper on the matter: http://dx.doi.org/10.4103%2F0972-124X.75909 And this: Difference between logit and probit models. Can you tell me when, generally, I should use ...
7
votes
2answers
562 views

Multivariate binary responses - advice on regression strategy

I would be grateful for advice on how to approach the following situation: I have a count variable X and four binary variables A, B, C, D. The count variable is the independent variable (it refers ...
2
votes
1answer
202 views

Univariate and multivariate regression

I found a research article Nurses’ reports of staffing adequacy and surgical site infections: A cross-sectional multi-centre study and I want to know the reason why they used regression for their ...
0
votes
2answers
151 views

Modeling influence of social impact with linear regression

Is it right to use linear regression to make a forecast based on social media impact? Suppose you have the next dataset (events), where time delta is amount of ...
1
vote
0answers
122 views

How to correct for selection bias caused by a prior version of a logistic GLM?

I've looked for information on this, but there are a lot of different angles to addressing selection bias I'm lost on how to proceed. Here is the basic scenario: We collect data on n customers. Of ...
1
vote
1answer
95 views

Using same variable as continuous and categorical

I have a pretty general question: I'm writing on my thesis and have some linear regressions that include a continuous variable. Is it okay to use the variables as continuous for the linear regression ...
1
vote
0answers
81 views

R or spss? For a conditional regression with 2 dependent variables?

Maybe you find my question a bit simple but I'm really confused as I'm not statistician. I have 2 SNPs (can be proposed 2 genes instead) that are related to a primary disease (like major depression), ...
3
votes
1answer
412 views

How to validate a ridge regression model?

How do I validate a ridge regression model? Presently I am using both the mean absolute error and the $R^2$ score from python's sci-kit learn, and am plotting the ...
4
votes
1answer
305 views

how many parameters can be estimated

First edition, Prof Harrell's REGRESSION MODELING STRATEGIES, section 8.2, "How many parameters can be estimated?", page 150, quoting "If predicting survival time were of major interest, we could ...
1
vote
1answer
31 views

Including a predictor that cannot be assessed in preverbal/non-verbal children

I'm building a prediction model (logistic regression, using R) with data obtained in children younger than 18 years. Twenty-five percent are less than 2 years old. One of my candidate predictors ...
1
vote
0answers
92 views

Dealing with "Don't know" and "missing" in the outcome variable

I want to use a generalized linear model for my analysis but the problem I am facing is that the outcome variable y which is categorical is not binomial since I have the following responses: Yes, No, ...
7
votes
1answer
7k views

Interpretation of calibration curve

I have a step-wise derived binary logistic regression model. I have used the calibrate(, bw=200, bw=TRUE) function in the rms ...
5
votes
3answers
3k views

Is there any limitation for the number of categories in Multinomial logistic regression?

It know it's very general question but I'm wondering what kind of issues should I expect if the number of categories in my dependent variable (or even in predictor variables) are more than for ...
19
votes
1answer
1k views

What does it mean to make the sample size a random variable?

Frank Harrell has started a blog (Statistical Thinking). In his premier post, he lists some key features of his statistical philosophy. Among other items, it includes: Make the sample size a ...
1
vote
1answer
322 views

Time-invariant variable problem in panel data

I have panel data with a time-invariant variable across firms. However, when I tested the xtsum after setting the panel data, there was very little (8.66e-17) within variance. I checked over and over ...
3
votes
0answers
168 views

Potential bias when the training set is more general than the testing set

I am using Logistic Regression (LR) to obtain Coronary Artery Disease CAD probability equation. The data set has 16 candidate predictors, all continuous. There are two groups, CAD patient group (70 ...
11
votes
2answers
1k views

How to start building a regression model when the most strongly associated predictor is binary

I have data set containing 365 observation of three variables namely pm, temp and rain. Now ...
1
vote
0answers
49 views

Use R to perform a logistic regression and cross-validate in a small sample?

I have a small sample (n=69, 35 with diseases, 34 without disease). I have around 20 variables that came up as significant using indiscriminate univariate regression. Of these 20 variables, several ...
2
votes
1answer
153 views

Estimation of treatment effect when there is an unknown and variable coverage of the population

I am not sure if I am using the correct terminology, something must be written about the following problem, but I cannot find it by searching. I am presently analyzing data about the effect of ...
1
vote
0answers
964 views

Why does gvlma() give contradictory results for linear model assumptions?

I have fitted a linear model and I am checking the assumptions for which I get this diagnostic plot: From this diagnostic plot there seems to be an increasing variance in the residuals for increasing ...