Skip to main content

Questions tagged [regression-strategies]

Regression Modeling Strategies

Filter by
Sorted by
Tagged with
2 votes
1 answer
58 views

Regression strategies for predicting a binomially distributed, count outcome: Poisson, Negative Binomial, and Logistic Models with Offsets

Data Description: I am working with a dataset of 100 hens, represented across four columns: ID: Numbers 1 through 100. Age: Each hen's age. EggCount: Number of eggs laid per hen, with a range from 0 ...
insan's user avatar
  • 145
0 votes
0 answers
11 views

How might I go about analyzing the affect that the number of attempts of something has on the failure rate?

I have a dataset where I know the number of attempts and number of failures for a large group of individuals. I have calculated the failure rate of each and I suspect that as there are more attempts ...
peeezy's user avatar
  • 1
3 votes
1 answer
120 views

How does non-collapsibility and the lack of an error term affect coefficients in regression

I have read from here that in nonlinear models such as the logit and Cox, because of a lack of an error term, coefficients may be biased (typically towards zero) when covariates are omitted; I see how ...
Geoff's user avatar
  • 741
1 vote
1 answer
55 views

Separate or One Large Multiple Regressions

I am investigating the relationship between scores on 3 questionnaires (SPQ,CAPS,PDI) and the effect of an experimental 'Condition' on performance (Correct/Incorrect). I have run the following ...
SilvaC's user avatar
  • 512
0 votes
0 answers
24 views

Modeling a chargeback rate as a function of an approval rate

I am working on a problem where I would like to estimate chargeback rates for online goods orders as a function of the proportion of orders that are approved (approval rate). The approval rate is a ...
alporter08's user avatar
5 votes
1 answer
69 views

With logistic regression, how does one choose a number of predictors when preregistering a study?

Harrell's Regression Modelling Strategies suggests that the number of predictors should not exceed $m/10$, $m/15$ or $m/20$.* For logistic regression $m$ is $\textrm{min}(n_1, n_2)$, where $n_1$ and $...
Mohan's user avatar
  • 939
0 votes
1 answer
213 views

If my logistic regression model is performing well, does it matter if my features don't pass the Box Tidwell Test?

I've built a logistic regression model for binary classification with a high F1 score, but when I run Box-Tidwell tests on continuous independent features/predictive variables, I find non-linearities ...
systems_engineer25's user avatar
0 votes
1 answer
297 views

Can you do a log transformation for excess kurtosis, or is that mainly used for skewness?

I am planning on doing a regression analysis on STATA on the financial performance of private equity funds. On my descriptive statistics, I saw higher levels of kurtosis and skewness. I decreased ...
Lucy's user avatar
  • 1
1 vote
0 answers
29 views

GVLMA give contradicting results on the same data compared with the inverse model

I was using GVLMA from R and this doc: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2820257/ But then I notice something weird with some models, working with the set of data, some tests of Y(X) can ...
Abs_0_'s user avatar
  • 11
1 vote
0 answers
51 views

Analysis with data from different sources

I have data from 3 different sources, measuring different variables for different samples taken from the same population (a country). All of the data is from country-wide studies and should be ...
ouroboro's user avatar
0 votes
0 answers
74 views

With what argument can I use splines method in a binary logistic regression analysis?

I want to run a binary logistic regression and I guess some variables have nonlinearities. So I want to use splines method to understand affect of each range in a continuous variable. When I 'guess' ...
Mostafa Ahmadi's user avatar
3 votes
0 answers
152 views

When fitting a generalised additive model, how to choose how much to smooth?

When fitting a GAM, is there a rule (of thumb) for deciding if $k$ (max number of degrees of freedom for a spline) is large enough or not? How much should edf be below $k'$? And is that an absolute ...
user avatar
2 votes
2 answers
205 views

Breaking the regression line into two pieces

My X & Y variables are associated like this below and I am trying to fit a simple linear regression model (y ~ x , data= df) , to estimate ...
Science11's user avatar
  • 577
1 vote
1 answer
31 views

pmin(x, 60) and subset regression [closed]

Is there any difference between y ~ pmin(x, 60) + sex , df = data and y ~ x + sex, df = subset(data, x <=60) If they are ...
Science11's user avatar
  • 577
3 votes
1 answer
313 views

How damaging to the analysis would it be to run probability validation (`rms::val.prob`) when calibration (`rms::calibrate`) is the correct action?

If I make a model that predicts probabilities (e.g., logistic regression or a neural network), I would like it to have the property that, when it predicts a probability of $p$, the event happens about ...
Dave's user avatar
  • 64.7k
1 vote
0 answers
114 views

How to compare the expected change in a exponential decay function with the expected change in a square root function?

I have two two datasets. Each contains two variables. There is one variable that is present in both datasets. When I plot each dataset to see the relationship between the two variables in each, I find ...
StatsScared's user avatar
  • 1,218
0 votes
2 answers
66 views

Change per capita, logarithmic change or non logarithmic change?

I am currently working with Covid-19 key figures such as registered cases of infection and death. My data is a panel dataset across time and municipalities in Denmark, the set consists of several ...
Jens Kramer's user avatar
5 votes
1 answer
563 views

Linear regression: highly-correlated features but not redundant

I have been working on a problem whose goal is to find the most accurate coefficient (from a physical point of view) that expresses the relationship between a feature and an output variable. The model ...
user10279396's user avatar
2 votes
1 answer
2k views

Walk through `rms::val.prob`

The val.prob function in the rms R package has similarities to the ...
Dave's user avatar
  • 64.7k
2 votes
1 answer
4k views

Walk through `rms::calibrate` for logistic regression

The calibrate function in the rms R package allows us to compare the probability values ...
Dave's user avatar
  • 64.7k
5 votes
1 answer
140 views

Can I interpret coefficients for "Year" as differences between years that are not explained by my predictors?

I am doing statistical analysis of a natural experiment that consists of multiple years of measurements. I have two independent variables that are physically related to ...
Felix Phl's user avatar
  • 299
1 vote
1 answer
598 views

Meaning of interaction with %ia% in rms? Three-way interaction?

In this very illustrative post on evaluating added value of predictors by Frank Harrell, he codes a logistic regression model as such: ...
Titorelli's user avatar
3 votes
1 answer
221 views

Estimating regression optimism using the bootstrap

I am estimating optimism bias in for example risk predictions. A method for doing that is described by Frank Harrell and implemented in the R package rms. I am ...
Danny's user avatar
  • 1,035
0 votes
1 answer
53 views

Articles about data analysis workflow

I am a non-statistician. I have to write a non-English article about my data analysis workflow for a particular epidemiological regression analysis that I conducted. The article will cover my workflow ...
ethan282712's user avatar
0 votes
1 answer
181 views

Effect size derived from LME longitudinal model: the statistical findings projected back down onto a group of people

I have been studying the change in a metric X with a linear mixed effect model. I have built this model in a multivariate setting, so I can see how each of my covariates (Time, sex, age) affect X. ...
Lili's user avatar
  • 220
1 vote
1 answer
19 views

Modelling strategies for analyzing an effect of a predictor through higher hierarchical level

What strategies can be considered when a predictor's direct effect can not be measured directly due to unmeasured confounding? However, data has a hierarchical structure (patients within regions) that ...
st4co4's user avatar
  • 2,217
1 vote
1 answer
103 views

Applying count models with rate responses

How do you apply count models to data which is count in nature, but a rate in reality? In such cases, r can handle this to a certain extent, depending on the model, but what is the correct way to ...
Ali's user avatar
  • 209
4 votes
0 answers
982 views

Boruta Algorithm for Logistic Regression?

Is it okay to use a Boruta algorithm to select features for a logistic regression? I read several sources, including the source package as well as this site explaining what Boruta does. My ...
Eric's user avatar
  • 41
2 votes
0 answers
500 views

Logit Regression and F-test: Can I apply the f statistic when variables are non-normal and the output is binary?

I want to do a univariate analysis on a set of variables to see which predict a binary outcome. I want to discard some of them before performing logistic regression. I am trying to understand if I can ...
Sapiens's user avatar
  • 224
4 votes
2 answers
2k views

External validation of a published Cox PH model

My aim is to externally validate a risk prediction model published in the medical literature that is based on a Cox regression model. I have a dataset with all the variables from the score. I read ...
b_surial's user avatar
  • 153
3 votes
1 answer
932 views

Brier score of calibrated probs is worse than non calibrated probs

The question is related to probability calibration and Brier score I have faced with the following issue. I have Random forest binary classifier and then I apply isotonic regression to calibration of ...
ABK's user avatar
  • 676
7 votes
2 answers
15k views

Handling missing data in logistic regression

I'm trying to do logistic regression, but I can't seem to get the results I want. I have 6 columns of data (one dependent and 5 independent binary variables) and about 100 rows. The problem with my ...
R.Rasheed's user avatar
7 votes
1 answer
828 views

probability calibration and Brier score

Assume that I have a binary classification problem. The outcome from classification I am mostly interested in is the well-calibrated probabilities. The first way to check this is the calibration plot (...
ABK's user avatar
  • 676
1 vote
0 answers
47 views

Determine minimum data to start with building model

We have developed a basic Regression framework where we try to build models for over 100 configs(stored in a file). To run : ...
CodeTry's user avatar
  • 133
3 votes
0 answers
137 views

OLR with rms: proportional odds assumption

I am fitting an ordinal logistic regression model with rms package. my data involves a three-level ordered outcome (see ...
Uri's user avatar
  • 71
5 votes
1 answer
730 views

When LASSO selects only parts of a categorical variable?

I want to use LASSO to construct a model and then run a logistic regression on the variables LASSO selects. However, LASSO selects only parts of some categorical variables that I put into it. Does ...
Paze's user avatar
  • 2,351
1 vote
2 answers
375 views

Excluding the effect of control variables in the assessment of a logistic regression model

I have a logistic regression model with ten independent variables of which two are included as controls. While their inclusion is necessary for correctly assessing the coefficients of the other ...
humperderp's user avatar
3 votes
0 answers
165 views

Using "moderately" correlated variables to select controls for a LASSO regression?

In medicine we often have a disease status as an outcome variable and a lot of independent variables in which we want to see if there is some connection. Traditionally, baseline characteristics such ...
Paze's user avatar
  • 2,351
0 votes
1 answer
167 views

How to validate Generalized Least Square model for longitudinal response

I have a dataset with body weights before and in the follow-up visits after surgery, for a group of patients with obesity. Our goal is to fit a model to predict weight loss throughout the follow-up. ...
Juan_Ramon Lacalle's user avatar
1 vote
0 answers
34 views

Combining mean differences in regressors and significant prediction and moderation in multiple regression

I am analyzing a survey about career choices with an eye toward demonstrating sex differences in: 1) the means for factors that might be related to career choice (e.g., differences by sex in ...
JOgawa's user avatar
  • 11
4 votes
1 answer
1k views

Conducting a subgroup analysis with regression modeling

I'm conducting a survival analysis using Cox Proportional Hazards regression to identify prognostic factors for cancer patients. My covariates include information such as age, sex, tumor location etc. ...
Tomas Bencomo's user avatar
1 vote
0 answers
22 views

how to choose the best logit model [duplicate]

I have two logit regression models with different AIC. I'm using R. my first model has significant variables and AIC 192.7436. And my second model has 1 non-significant variables but with smaller AIC ...
Fatma Hanani's user avatar
1 vote
0 answers
23 views

I have a confusion between using 4 different general linear models and 1 singular ones. I have provided with the codes and outputs

I want to check the effect on mass of crickets, I have a fixed linear effect (AltitudeAge), fixed quadratic effect (AltitudeAge^2), random effects (Nymph IDs, population and the incubators they are ...
eggde43's user avatar
  • 11
1 vote
1 answer
3k views

Number of Covariates in Cox PH Model and Overfitting

I have a small time to event dataset (N=20) where patients are given one of two drugs (drug) at varying doses (...
Tomas Bencomo's user avatar
2 votes
1 answer
410 views

Fitting model again after variable selection

This question has been asked quite a bit in other contexts (doing LASSO then OLS on selected variables for example), but I'm unsure about how to proceed for this case. Suppose I have a set of 50 ...
Joel H's user avatar
  • 98
5 votes
1 answer
2k views

Problems from having too many interactions in a regression?

Excluding the 'dummy variable trap', are the problems from including too many interaction terms in a regression any different from the problems of including too many continuous or binary variables in ...
StatsScared's user avatar
  • 1,218
23 votes
2 answers
2k views

Does LASSO suffer from the same problems stepwise regression does?

Stepwise algorithmic variable-selection methods tend to select for models which bias more or less every estimate in regression models ($\beta$s and their SEs, p-values, F statistics, etc.), and are ...
Alexis's user avatar
  • 30.3k
1 vote
0 answers
136 views

Kaplan Meier Diagnostic Utility

I'm trying to understand a paper that claims to have identified a gene expression signature that can distinguish primary from metastatic tumors. The authors stratify their data into patients with and ...
Tomas Bencomo's user avatar
1 vote
0 answers
542 views

Logistic regression with repeated mesures and unique outcome

I have one independent continuous and time-dependent variable X, repeatedly measured (from 1 to 4 times) in different patients during some period of time. My dependent variable Y is binary and is ...
Esculape's user avatar
3 votes
1 answer
3k views

Interpretting Cox Regression ANOVA

I'm having difficulty interpreting the results from anova() in the rms package. My confusion arises from what information the <...
Tomas Bencomo's user avatar

1
2 3 4 5
7