All Questions

Filter by
Sorted by
Tagged with
0
votes
0answers
3 views

Should I use Pearson or Spearman correlation (and is it even correct)?

so I'm doing the following: for a research paper at university I want to calculate the correlation between top grossing ranks and rating scores of mobile apps. I pulled the data from an analytics site ...
0
votes
0answers
2 views

Can I use Cumulative Logistic Regression in spite of violation of proportional odds assumption?

I have an ordinal outcome variable with the following levels (unlikely, somewhat likely, very likely) that I want to tests in an Adjusted Regression Model with about 20 predictor variables (including ...
0
votes
0answers
4 views

Feature Selection when LASSO sets all coefficients to zero

I am trying to select features for a data set that contains what food people have eaten with a result of BMI. However, since I want to see what foods impact BMI most I am trying to use various forms ...
0
votes
0answers
2 views

Convolution of a multivariate Gaussian and multivariate exponential distribution

I have derived the convolution for a multivariate Gaussian distribution with a multivariate exponential distribution. By having such a distribution in place, one can have the best of both worlds: ...
0
votes
0answers
5 views

F-distribution likelihood interval, does it have to be symetric?

F-distribution : https://en.wikipedia.org/wiki/F-distribution I was told that the confidence interval for the $F$-function is a symetric interval for instance : $$ [ F_{1 - \alpha / 2, n-1 , m-1} \...
0
votes
0answers
4 views

Is my Data stationary? KPSS, ADF Tests and ACF

I already differenced my Data by 1 and i am not sure whether my Data is now stationary or not. I perfomed an KPSS and ADF test in order to help me decide if it is. I think it is stationary but im not ...
0
votes
0answers
7 views

Model selection based on training data

Suppose I have bunch of models and I need to figure out the best model. I calculate the likelihood for each model on the training data set. Why cannot I use likelihood from the training data set for ...
0
votes
0answers
5 views

Determining the percentage of males and females employed by the company using mean of salaries

The mean monthly salary paid to all employees was 1500. The average monthly salaries paid to male and female employees were 1560 and 1260 respectively. Determine the percentage of males and females ...
0
votes
0answers
13 views

how to get best fit line when we have data on vertical line?

I started learning Linear Regression and I was solving this problem. When i draw scatter plot between independent variable and dependent variable, i get vertical lines. I have 0.5M sample data. X-axis ...
0
votes
0answers
7 views

What's the interpretation of the regression between residuals and fitted values?

I'm using the teengamb dataset, from the faraway package in R. I will explain my line of thought, so if I'm mistaken at some ...
0
votes
0answers
6 views

Labels in Out-of-time validation

I have a binary classification problem and I use out-of-time validation to validate my models. My question is regarding the label. There is a lag in identifying the correct label. Simplified example: ...
3
votes
1answer
11 views

Loss function to widen distribution of predictions

I'm working on a project where my business counterparts want my model's distribution of continuous predictions to more closely represent the distribution of actual values. Below is an example of what ...
0
votes
0answers
11 views

Evaluation of variance components - mixed models

How can I evaluate if the variance components of a nonlinear mixed model make sense (assuming or not assuming treatments factor)? For instance, if I am assuming an unstructured variance-covariance ...
0
votes
0answers
5 views

randomGLM Error: in singleBagIteration(bag, verbose = verbose) : task 1 failed - “'x' must be numeric”

I am trying to build a predictive model for a multiclass outcome. I am using randomGLM() function. Here is my code (I took the example from the package ...
0
votes
1answer
9 views

Are there good alternatives to Cochran's Q test for heterogeneity in meta-analysis?

I am not a statistician so please bear with me. We used the Q test to determine heterogeneity with p-values <0.10. A highly recognised statistician commented that the Q test has low power and that ...
0
votes
0answers
5 views

Should deviation coded predictor be classified as factor or numerical for mixed model?

I have an ordinal mixed model with several multilevel categorical factors. These factors are deviation coded as there is no hierarchical structure but rather represents different groups such as sex ...
0
votes
0answers
7 views

MLE for t distribution parameters using EM in R

I have a set of randomized data from a t-distribution and I want to estimate the parameters using the MLE. I know this can be done using EM, however, I am stuck as to how to compute that in R. I have ...
0
votes
0answers
4 views

Would common method bias result if I measure two or more independent variables to be included in a same regression analysis?

From what I read, it seems that common method bias would only result if both independent and dependent variable are measured with the same way (e.g. via survey of the same people). But what about the ...
0
votes
0answers
15 views

Little to no difference between OLS, FGLS, and WLS coefficients

I am running a regression of crop yields on Carbon Dioxide and temperature. I have run through multiple specifications (quadratic and interaction terms) and settled on a simple linear model. I have ...
0
votes
0answers
3 views

What are appropriate methods for preparing categorical features for recurrent networks?

Given a 1D sequential categorical input variable, e.g. [rainy, sunny, rainy, cloudy, cloudy], with a small domain {rain, sunny, cloudy}, what encoding methods (e.g. one-hot, dummy, binary) and what ...
0
votes
0answers
4 views

Can I compare the AIC of gamm models with s(x) and s(x,by=z) predictors?

I am fitting a set of generalized additive mixed models using gamm or gamm4 packages for R, and in particular want to compare the AIC of two models of these forms: mod1 <- gamm(y~s(x), random=~(1|...
0
votes
0answers
5 views

Can I use a capture-recapture design to estimate market size in a B2B SaaS environment?

I am looking to provide statistically sound market size (number of businesses) and market share (percent of total businesses) estimates for a B2B SaaS business in an environment where data is readily ...
0
votes
0answers
8 views

Is a Log-Log LM the same as an Identity-Log GLM with a log-link?

Intuitively, it seems like they would be doing the same thing, but I am getting different results between the two models. What is going on under the hood that would result in different coefficient ...
0
votes
0answers
5 views

Time units as features vs individual records

I am working on a project similar to predict likelihood of a student completing an online course given the performance to a given date. The approach taken so far is very similar to this https://...
0
votes
0answers
3 views

Depthwise Separable Convolution

I understand that depthwise separable convolutions are computationally cheaper than normal convolutions, but are the two operations equivalent in its outcome? Given the same batch of images and ...
1
vote
1answer
9 views

Group comparison for small sample with repeated measured data

I have very small dataset. When I conducted rank sum test between the groups (A vs B; A vs C), I found no significant results, but just trends level difference. Is there any alternative statistical ...
1
vote
1answer
31 views

Using age categories in regression

I used multiple age categories (under18, 18-35, 36-44 etc.) in my survey. So I recoded each interval as a dummy variable. I know it's not 100% correct to use something like age as categorical ...
0
votes
0answers
7 views

How to conduct meta-analysis for thousands of genes at one time using R package “metafor”

I'm trying to do a meta-analysis on many genes using R package "metafor", I know how to do it one gene at a time but it would be ridiculous to do so for thousands of genes. Could somebody help me out ...
0
votes
0answers
10 views

Application of circular data in many field

reserach about circular data in medicine espacially reserach in eye and reserach of application in social science and Geology and Earth science . Please, can put any website in this filed .
0
votes
0answers
6 views

Why several studies change logical paired/unpaired assumption from omnibus to post-hoc?

There are several studies of highly-referenced researchers where a paired post-hoc test (e.g. Wilcoxon signed-rank) follows the omnibus unpaired test (e.g. Kruskal-Wallis): From Huang et. al "THE ...
0
votes
0answers
6 views

Can I learn something from data that is not very well sorted in the domain?

I have an application in which I produced some parts varying 4 parameters (in1 ... in4), and I'd like to analyze their influence in 2 output parameters (A and B). The problem is that the parameters ...
0
votes
0answers
5 views

How can analytical weights be used together with sampling weights?

I run a linear probability model (LPM) on survey data, which contains sampling weights. Say the predicted probabilities from the OLS regression are $\hat p_i$. Heteroskedasticity would make me ...
0
votes
0answers
11 views

Is there a way to create a Kaplan Meier curve with multiple events, rather than one occurrence?

Thanks in advance for any advice regarding my question. I have KM curves created for probability of (not) getting an infection within our patient population. I am wondering if there is a way to ...
0
votes
0answers
17 views

Probability distribution with percentage values where the pdf does not exceed 1?

If I want to estimate the probability of a value between 0 and 1 appearing, given historical data, then how would I set up a probability distribution, such that the y-axis shows the probability of ...
1
vote
0answers
7 views

Is repeated measures ANOVA/paired t-test the correct test?

I've been doing some reading on the correct test to use for a project I'm working on, and I've found some recommendations but I haven't yet found something that addresses the exact kind of study I'm ...
0
votes
0answers
3 views

How to fit Gaussian Process with High-Dimensional Data Sets in R-INLA?

I am new to R-INLA and it really looks like a cool and versatile framework. I wonder if I could use INLA to fit high dimensional Gaussian Process (with 8-20 covariates), i.e.: $${f}(\textbf{x}) \sim \...
0
votes
0answers
7 views

Use of count in Stata

I am using the data set of Verboven cars1.dta Download here. Then, I use egen yearcountry=group(year country), label and ...
2
votes
1answer
31 views

What do I do when my model prediction exceeds the limit of 100%?

I have plotted the effects of a model glm( A ~ B, family=poisson, data=data) both with the sjplot R package and with the ...
0
votes
0answers
6 views

Marginal effects in multinomial logit in R

I am trying to calculate average marginal effects (dF/dx) for a multinomial logit model in R. Package mfx provides the solution only for binomial (and not the multinomial) model. Is there a package or ...
1
vote
1answer
18 views

Are parametric tests only subject to ratio and interval scale measurements?

I had doubts on the above flow chart based on measurement scales which I found in a published article. Can anyone correct m please and clear my doubts as I believe non parametric tests are not ...
0
votes
0answers
14 views

Is IV really invariant to the number of bins?

I read several times that "the IV formula is fairly invariant to the number of bins". This is no very intuitive to me, because: IV is a sum of positive numbers (one number per bucket). Therefore, if ...
0
votes
0answers
13 views

Measure the distance between two probability transition matrices

I have a probability transition matrix $P$ that contains some values very close to zero. I want to sparsify this matrix by taking the k largest values for each row and setting the others to zero. For ...
0
votes
0answers
15 views

Testing the consistency of two sets of measurements

Suppose I'm measuring a quantity $X$ twice, $x_{1}\pm\sigma_{1}$ and $x_{2}\pm\sigma_{2}$. $\sigma_{1}$ and $\sigma_{2}$ are in general not equal to one another. If I want to know if the two values ...
1
vote
0answers
4 views

scikit-learn IsolationForest no variance feature

I'm using IsolationForest algorithm in order to detect anomalies in my data and to use this model to detect future anomalies in new rows and came across a few questions: Is the model good for ...
0
votes
0answers
7 views

Why does the test statistic of a 2 Sample K-S test change drastically when noise is added to samples?

I believe I have gone through all 2 Sample K-S test related posts here, but I could not see the answer to my problem. At this point, I don't know whether the problem is related to a bug in the ...
3
votes
1answer
34 views

p-values in multiple linear regression depend on number of predictors?

So I am wondering if there usually is any "correction" being made to p-Values (of single predictors) if you add more predictors to the model? Does anybody know how SPSS handles this? I am well aware ...
-1
votes
0answers
13 views

Product of normal random variables and exponential random variables is exponential random variables?

How we can prove analytically, the Product of normal random variables and exponential random variables is exponential random variables?
0
votes
0answers
10 views

Weighted average Sufficient Statistic

I have the following question about sufficient statistic: suppose that you have N agents indexed by $i$ and $j$ each starting with a noisy signal $x_i^0 = \theta + \epsilon_i$ where $\theta$ is a ...
0
votes
1answer
18 views

Does it make sense to use correlation between your predicted value and your true value as a metric for overffiting?

Does it make sense to get the correlation between your predicted value (y_hat) and your true value (y) as a metric for overfitting for your training data? My logic is that if you have a predicted ...
0
votes
1answer
18 views

Understanding ANOVA to compare Mixed Model with a GzLM

I'm having a hard time understanding how can I compare a GLM with a GLMM, knowing that I probably can't compare their AIC as glmer from lme4 probably computes the maximum likelihood differently from ...

15 30 50 per page