# Tag Info

0

IMHO it's a good idea to publish both results. This is because you should do you uttermost best to explain the situation to the reader. I think that a significant result is not the full story. To that point, this model isn't beyond criticism. Here you assume a fixed linear trend for the proportion of males (presumably with a logistic link). Fixed, the ...

2

How precisely you should report your model depends on your context - we can't help you here. The size of the groups is usually not the culprit, it's simply the different relationship between the predictor and the response for different groups, specifically an intercept that differs between groups. If you describe this, I would say you are doing the right ...

2

The curve is valid. The most commonly seen ROC curves have a curve above the diagonal, which indicates that large predicted scores are associated with the label, and low predicted scores are not associated with the label. The diagonal, i.e. the line $\text{TPR}=\text{FPR}$ of a ROC curve corresponds to the model that has a completely random relationship ...

1

It is rather rare to find tabular results in a scientific journal that do not report the study's total sample size $N$. A reviewer may raise an eyebrow if the reporting of your $N$ is entirely absent. A journal's editorial board might even require it in the presentation of tabular figures. In your case, a 'data scrape' of approximately 17.5 million ...

3

What you have is a (labeled) scatterplot. To a first approximation, what you can learn from this is what you can learn from any scatterplot. Probably what would help is to think about what PCA does. The text below is copied from my answer here: Put simply, PCA (as most typically run) creates a new coordinate system by: shifting the origin to the centroid ...

0

There is a valid point in the comments about degrees of freedom in the mixed model. However, I suspect that this knowledge will lead you towards an answer, and it’s too long for a comment. The F-test can test groups of variables, such as dog/cat/horse, which you would represent with $(0,0)$, $(1,0)$, and $(0,1)$. To be consistent with what they were doing ...

1

For Question 1, your binary logistic regression model would be expresed like this: log(odds of having at least one girlfriend) = beta_0 + beta_height * height + beta_color * color So you would interpret beta_height as follows: Among subjects in the target population with the same value of color, each extra cm of height is associated with beta_height ...

0

When writing it up, you should interpret the model as follows. Your variable "FIRST_cat" was significant, you can reject the null hypothesis that it is equal to zero. If that variable is "high" it has a positive effect, otherwise a negative effect. Furthermore, that means that the probability of "InsomniaT3" being "high&...

0

As you are using factors, glm automatically uses dummy variables to estimate your coefficients. It created 2 dummies: one for FIRST_cat = "High", named FIRST_catHigh, and other for FIRST_cat = "Low", but as you can't add both on the model (because of multicolinearity problems), it uses only one. The coefficient associated with the ...

0

Bear with me here, I will explain a case where the variable is not binary as I think you will understand better how R behaves with binary variables. Let me start with a linear regression model $\hat{y} = \beta_0 + x_1\hat\beta_1 + x_2\hat\beta_2$ where $x_1$ is a numerical variable and can take any real valued number $x_2$ is a categorical variable, for ...

2

With only 31 observations in your data set, you don't have balance. As the aov manual page says: aov is designed for balanced designs, and the results can be hard to interpret without balance... Your data set is specifically missing a 32nd case for which B=1, A=4, and S=1. If you provide such a case along with a corresponding value of Y, restoring balance, ...

0

Nullity correlation ranges from -1 to 1. Zero indicates the variables do not have any correlation. -1 indicates a strong negative correlation where +1 indicates a strong positive correlation (think of it as the direction of slope)

0

Whether to use Poisson or Gamma regression shouldn't depend on whether the data are integer-valued, that is a minor consideration. In the quasi-GLM framework you can use Poisson regression with non-integer data. The key difference between Gamma and Poisson regression is how the mean/variance relationship is encoded in the model. The Poisson approach models ...

2

The correct and complete interpretation for b2 is as follows: Among US beneficiaries with the same body mass index (bmi), those who live in the northwest region of the US have 16.51% lower odds of incurring charges of 10000 dollars or more than those who live in the northeast region of the US. Notice the use of plural for odds and also the fact that we are ...

0

To read the plot: The variable names are on the diagonal plots, llok at the bottom row: All the four plots there has acceleration as the horizontal, x variable. For the y variable, go up to the diagonal and find the variable name there. As for the correlations, you can just compute them all! See Can I analyze or model a conditional correlation? for code ...

2

The key here is that in a saturated factor model like what you have above, "with southeast and southwest regions are fixed" is a superfluous. If you are in the northwest region, you are by definition not in the southeast or southwest region. I find that it helps clarify my thinking to recognize that in this setting where you are just regressing on ...

2

Note here that the reference category for the region variable is northeast. Therefore, assuming that you are using the default contrast coding: (Intercept) 13406.4 means that the outcome has an expected value of 13406.4 for the northeast region. regionnorthwest -988.8 means that the northwest region has an expected value for charges of ...

1

This transformation results in something very close to least squares residuals. At first step you can regress vextor $\pmb x$ on the set of $m$ dummies. In such case all other than $\pmb x$ columns of the $\pmb A^T$matrix would be OLS matrix $\pmb X$. When performing such action the plot of negative residuals will be very similar to your transformation ...

Top 50 recent answers are included