Questions tagged [regression]
Techniques for analyzing the relationship between one (or more) "dependent" variables and "independent" variables.
30,254
questions
0
votes
0
answers
19
views
Is there any value in models that have a larger out of sample RMSE than a standard deviation?
I am predicting y values from x values using various regression models, elastic net and partial least squares regression (PLSR). To quantify performance of models we utilize root mean squared error (...
- 1
2
votes
2
answers
82
views
Recreate `lm` Categorical Regression
Consider the code, which contains regression using lm of two categorical and one continuous variables without interaction using data from the correct model:
...
- 552
0
votes
1
answer
24
views
Is differential follow up periods in survival analysis a problem?
I'm looking to compare mortality post-operatively following two different surgical techniques. due to technological advances, one of these surgical techniques was only performed around five years ...
- 1
1
vote
0
answers
31
views
Gaussian linear model marginal likelihood under g-prior
Consider a Gaussian linear model with an $ n \times 1 $ outcome vector $ y $ and an $ n \times p $ matrix of centered predictors $ X $:
$ y = \iota\alpha + X\beta + \varepsilon \quad \quad \varepsilon ...
- 720
4
votes
2
answers
131
views
Does artificially balancing outcomes in regression lead to poor calibration? If so, how to show the poor calibration?
In "classification" problems, it is common for there to be unbalanced-classes. To combat what appears to be a problem (though I would argue that it usually is not a problem), it is common ...
- 65.6k
-1
votes
1
answer
32
views
Can I use a simpler model (small R²) to get more statistically relevant results?
I am new to this kind of stuff.
I am currently writing a bachelor thesis that uses data from the European Social Survey (round 7). 4 important questions that I'm using were not answered by every ...
2
votes
0
answers
92
views
Poisson two-way ANOVA test for Before and After treatment (including a control)
I am having trouble to determine what type of test to use in this experimental design:
My data are like this:
I got 10 treatment sites and 4 control sites
Control sites are outside an area in which a ...
- 164
0
votes
1
answer
27
views
Should I Use Regularization in Univariate Logistic Regression for Diagnostic Methods Comparison?
I am comparing two diagnostic methods, Method 1 and Method 2, where Method 2 is considered the gold standard. I am using Method 1 to predict the Method 2 using logistic regression. My dataset contains ...
1
vote
1
answer
28
views
What are some issues with using robust standard errors when you do not need to?
I've been reading the web about this. Some text books mention one line in passing that it can violate some assumptions but never really say which ones.
When googling I seem to get into a rabbit hole ...
- 362
7
votes
3
answers
459
views
I think standard deviation of y is related to size of x. How do I create a model for this / test this?
I have a sample of data $(x_i, y_i)$. I hypothesize that $y_i$ is not dependent on $x_i$, but the standard deviation of $y_i$ depends on $x_i$
More concretely, say I assume $\textrm{Var}(y_i | x_i) = ...
- 716
0
votes
1
answer
25
views
Predict change in dependent variable over three time points using independent variable at baseline
I want to predict the change in my dependent variable (DV) over time using an independent variable (measured only at baseline). My DV was measured at three points in time. The easiest way would be to ...
- 213
1
vote
0
answers
20
views
Is covariance estimation via the inverse hessian method generalizable (or possible) for loss functions other than least squares?
I know from other resources such as here that the scaled inverse hessian of your least squares loss can be used to estimate your model's parameter uncertainty (specifcally, covariance), but I can't ...
- 11
1
vote
0
answers
22
views
Why is the threshold term incorporated into the weight vector in linear classifiers?
In the context of linear classifiers, such as the perceptron or logistic regression, I understand that the decision boundary is defined by a linear combination of input features and weights, plus a ...
5
votes
2
answers
399
views
Standard negative binomial regression when counts are mainly zeros?
This question must have been asked many times before but I can't find an answer.
I'm getting very confused about when to use a zero-inflated negative binomial regression vs standard negative binomial ...
- 493
0
votes
0
answers
12
views
Bayesian Mediation Analysis
I have:
1 binary outcome (0, 1)
1 continuous quantitative mediator
1 continuous quantitative predictor
I would like to compute Bayesian mediation analysis with Liu et al. (2023) method. The formula ...
- 41
1
vote
1
answer
40
views
Optimal sequence analyses and hypothesis testing
I am working with data in R using the TraMineR package. I am fairly new to this type of analysis, so please bear with me.
I have a list of unique states (Ex: "NN", "s2s", "...
- 11
0
votes
1
answer
32
views
Elasticity from differenced log-level regression
I have the regression
$ \Delta \ln Y_i = \alpha + \beta \Delta X_i + \varepsilon_i $
where $\Delta \ln Y_i = \ln Y_{t,i} - \ln Y_{t-1,i}$ and $\Delta X_i = ((X_{t,i} - X_{t-1,i}) / T_{t-1,i} ) \cdot ...
- 363
0
votes
0
answers
21
views
Does the linear regression fit follow the inverse relationship? [duplicate]
I fitted a X =logA, Y=logB with a weighted linear regression and I got the result as log B =(0.53 $\pm$ 0.054)logA + (17.41 $\pm$ 1.7). When I did the fit with X=logB, Y=logA, I expect the ...
0
votes
0
answers
47
views
Are $X_i \varepsilon_i$ iid?
Take:
$X_i , \ i = 1, ... , n $ iid.
$\varepsilon_i , \ i = 1, ... ,n$ also iid.
$X_i \not \perp \varepsilon_j$ (they are not necessarily independent)
Are $X_i \varepsilon_i$ iid ?
- 1
1
vote
1
answer
42
views
My data is clustered in space, how can I account for that in a GAMM?
I am modelling whether a bird nest is build or not based on the date it was built, the geographical coordinates where the nest was built, and using the year when the nest was built as random effect - ...
- 21
0
votes
0
answers
23
views
Compare effect of two coefficients in logit model
I have a logit model with multiple independents qualitatives variables (A, B, C), and I would like to compare the probability of success between individuals who have a certain profile.
For instance, ...
- 109
1
vote
1
answer
15
views
Analyses for a vocabulary profiler data
I am writing a paper which uses data from vocabulary profiler (a tool that tokenizes and analyzes a text, output is something like "This text consists of 20% of Level X vocabulary, 20% of Level Y ...
- 13
4
votes
1
answer
125
views
Validity of AIC When Comparing Models with Varying Dispersion Parameters
I'm currently making a binomial model with a logit link, which is parameterised as a quasibinomial since I'm allowing it to calculate the dispersion parameter. I was wondering, since changes to the ...
- 95
3
votes
2
answers
64
views
Justifying Residual Histograms and QQ Plots for Linear Regression
Conceptually, I am having a hard time as to why we consider the quantile-quantile plot for linear regression diagonistics, and I cannot seem to get a clear answer after searching extensively.
The ...
- 161
2
votes
2
answers
51
views
What methods to use in pre and post testing?
I am currently running a pre- and post-intervention study. My hypotheses are if the intervention has an effect and if personality traits moderate the effect of the intervention.
I have an initial ...
- 23
7
votes
2
answers
305
views
How much dispersion is too much for quasipoisson regression?
Quasipoisson regression goes beyond standard poisson regression in taking into account overdispersion (whereby the dependent variable's variance is much greater than its mean). This is explained at ...
- 12.8k
2
votes
1
answer
66
views
Multilevel Model Residuals Scatterplot Assumptions
I am conducting multilevel modelling (MLM) in SPSS (mixed modeling) to analyze cross-sectional repeated measures data. One of my dependent variables is a survey question scaled 1 to 10, which ...
- 135
8
votes
1
answer
654
views
How should I analyse TV episode popularity while accounting for time?
I have a toy dataset which contains TV episodes, the date they were released and how many streams they have received:
episode
date
streams
The Shadow Realm
2024-03-08
5987
The Forbidden Tome
2024-...
- 197
0
votes
3
answers
40
views
Compare the proportion of multiple diseases between 2 groups
I'm conducting an analysis on data derived from two groups subjected to different environmental conditions. Here's a brief overview:
Group A: 750 individuals exposed to smoke.
Group B: 1500 ...
- 27
1
vote
1
answer
70
views
Which statistical test would you recommend for this experimental design?
I measured the percentage of individuals that crossed an averisve barrier at six different time points for four different genotypes.
The barrier is made out of an aversive substance that the control ...
- 23
0
votes
0
answers
33
views
Regression with noises in X. Should I use the unbiased estimator or the OLS estimator for forecasting?
I am working with a dataset that includes variables $Y$ and $X$. I assume that
$$ Y = \beta X + \epsilon $$
satisfies all the assumptions of OLS. Based on industry knowledge, I know that theoretically ...
- 235
0
votes
0
answers
32
views
Simple OLS Regression Assumptions
Forgive me; I am new to learning statistics. I am trying to understand OLS simple linear regression better and gain an intuition for the associated assumptions. Upon my research, I have identified ...
- 161
0
votes
0
answers
13
views
How to perform model comparison based on multinom( ) function of nnet package in R? [duplicate]
My independent variables are gender and sequence, and the dependent variable is intervention (including 3 intervention methods). I established a multinomial logistic regression model to examine the ...
- 23
2
votes
1
answer
43
views
How to perform model comparison based on multinom( ) function of nnet package in R?
My independent variables are gender and sequence, and the dependent variable is intervention (including 3 intervention methods). I established a multinomial logistic regression model to examine the ...
- 23
3
votes
1
answer
277
views
A model suffering from omitted variable bias can be said to be unidentified?
If my regression model
$$
y_i = \alpha + \beta x_i + \epsilon_i
$$
suffers from OVB the error contains one variable which we assume correlated with
$$
\epsilon_i = \gamma w_i + u_i
$$
my estimate of $\...
- 539
0
votes
0
answers
27
views
How to choose a feature as an offset in GLM Binary Regression
Is there any statistical test to determine that Feature A should be an offset instead of Feature B and Feature C.
From what I understand, if in insurance modelling, usually, offset has a connection ...
0
votes
1
answer
53
views
Dealing with categorical variables on cox regression
I am trying to fit a Cox regression model to my time-to event data and have a categorical variable with 5 different levels. If I don't leave one of the levels out, then I will have multicollinearity, ...
- 57
0
votes
0
answers
23
views
Effects Contrast Matrix for Interaction with Multiple Grand Means
I want to include an interaction between two factors: area and house_type, in my GAM (MGCV), I hope to make a sum-to-zero ...
- 76
0
votes
0
answers
21
views
I am confused about the formula SST = SSE + SSR [duplicate]
I dont know why in this case this formula does not hold.. Can someone explain?
0
votes
0
answers
20
views
Why residual term $\epsilon$ is not included in logistic regression specification? [duplicate]
I have a simple question, which bothers me. In the logistic regression context, the odds are defined as:
$$\frac{p(X)}{1-p(X)} = e^{\beta_{0}+\beta_{1}X}.$$
I was wondering, why in the literature the ...
- 487
0
votes
0
answers
22
views
Multivariate t-test equivalent - control versus intervention
Suppose there were two groups characterised by the same dependent and independent variables.
Dependent variable = time to treatment initiation - continuous.
Independent variable = point of care test -...
- 244
2
votes
1
answer
57
views
Regression with dependent observations of only one individual
Last week, I received a task to plan an analysis that my team wishes to perform.
My objective is to measure if one physician agrees with the outputs that a certain tool generates for a set of N ...
- 53
4
votes
1
answer
275
views
Can I include a variable related to the outcome variable into statistical analysis?
My research question is about the contact patterns during the pandemic and what characteristics of people who contacted more person during the national lock down.
The outcome variable is a variable ...
- 171
1
vote
1
answer
70
views
Date as random effect in mixed model strongly changes coefficient estimates
I am struggling with the structure of a mixed model that I run with lme4.
I have measured a behaviour (let's say reaction time) and another variable that might impact it (let's call it "mood, ...
- 33
1
vote
1
answer
26
views
Subjective confidence as weights in regression models
I have data where subjects rate a quantity on a certain scale ($y$) but also add their subjective confidence as to how sure they are in their choice ($w$). My initial thought was to add $w$ as weights ...
- 3,560
2
votes
1
answer
34
views
When the direction of causation can plausiblely run in either direction, is there a good stat approach to distinguishing relative strength?
I am interested in an instance where it is plausible that ideology influences economic outcome, and also that economic outcomes influence ideology. Assuming that I have good and relavent measures of ...
- 3,187
1
vote
1
answer
55
views
Is it reasonable to split the data and then perform the regression?
There's a store that sells ballpoint pens. I have obtained data on each transaction from this store, including transaction IDs, colors, sell number per transaction, total prices, etc. By dividing the ...
- 11
1
vote
1
answer
21
views
AR(p) model in R not fitting data [closed]
I have a set of data that I am trying to model with a simple AR(p) model. I've run a Dickey Fuller test on unit root stationarity and reject the null.
However, when I run a simple ar command in R I ...
- 539
0
votes
0
answers
18
views
Measuring treatment heterogeneity
This follows on a first question posted here
I am looking at the effect of information on local prices and choices to move to certain cities. I think that this information will have a positive effect ...
- 317
7
votes
1
answer
169
views
Interpreting interaction term
Suppose I am interested in estimating the effect of a treatment (vs a control), fully and conditional on different levels (say, 3) of a baseline caracteristic (say, income level). I would have a ...
- 317